Human Papillomaviruses: Methods and Protocols (Methods in Molecular Biology)

M E T H O D S I N M O L E C U L A R M E D I C I N E TM Human Papillomaviruses Methods and Protocols Edited by Clare Da...

Author: Clare Davy | John Doorbar

147 downloads 1236 Views 7MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

M E T H O D S I N M O L E C U L A R M E D I C I N E TM

Human Papillomaviruses Methods and Protocols Edited by

Clare Davy John Doorbar

Human Papillomaviruses

M E T H O D S I N M O L E C U L A R M E D I C I N E™

John M. Walker, SERIES EDITOR 124. Magnetic Resonance Imaging: Methods and Biological Applications, edited by Pottumarthi V. Prasadi, 2006 123. Marijuana and Cannabinoid Research: Methods and Protocols, edited by Emmanuel S. Onaivi, 2006 122. Placenta Research Methods and Protocols: Volume 2, edited by Michael J. Soares and Joan S. Hunt, 2006 121. Placenta Research Methods and Protocols: Volume 1, edited by Michael J. Soares and Joan S. Hunt, 2006 120. Breast Cancer Research Protocols, edited by Susan A. Brooks and Adrian Harris, 2005 119. Human Papillomaviruses: Methods and Protocols, edited by Clare Davy and John Doorbar, 2005 118. Antifungal Agents: Methods and Protocols, edited by Erika J. Ernst and P. David Rogers, 2005 117. Fibrosis Research: Methods and Protocols, edited by John Varga, David A. Brenner, and Sem H. Phan, 2005 116. Inteferon Methods and Protocols, edited by Daniel J. J. Carr, 2005 115. Lymphoma: Methods and Protocols, edited by Timothy Illidge and Peter W. M. Johnson, 2005 114. Microarrays in Clinical Diagnostics, edited by Thomas O. Joos and Paolo Fortina, 2005 113. Multiple Myeloma: Methods and Protocols, edited by Ross D. Brown and P. Joy Ho, 2005 112. Molecular Cardiology: Methods and Protocols, edited by Zhongjie Sun, 2005 111. Chemosensitivity: Volume 2, In Vivo Models, Imaging, and Molecular Regulators, edited by Rosalyn D. Blumethal, 2005 110. Chemosensitivity: Volume 1, In Vitro Assays, edited by Rosalyn D. Blumethal, 2005 109. Adoptive Immunotherapy: Methods and Protocols, edited by Burkhard Ludewig and Matthias W. Hoffman, 2005 108. Hypertension: Methods and Protocols, edited by Jérôme P. Fennell and Andrew H. Baker, 2005 107. Human Cell Culture Protocols, Second Edition, edited by Joanna Picot, 2005 106. Antisense Therapeutics, Second Edition, edited by M. Ian Phillips, 2005 105. Developmental Hematopoiesis: Methods and Protocols, edited by Margaret H. Baron, 2005 104. Stroke Genomics: Methods and Reviews, edited by Simon J. Read and David Virley, 2004

103. Pancreatic Cancer: Methods and Protocols, edited by Gloria H. Su, 2004 102. Autoimmunity: Methods and Protocols, edited by Andras Perl, 2004 101. Cartilage and Osteoarthritis: Volume 2, Structure and In Vivo Analysis, edited by Frédéric De Ceuninck, Massimo Sabatini, and Philippe Pastoureau, 2004 100. Cartilage and Osteoarthritis: Volume 1, Cellular and Molecular Tools, edited by Massimo Sabatini, Philippe Pastoureau, and Frédéric De Ceuninck, 2004 99. Pain Research: Methods and Protocols, edited by David Z. Luo, 2004 98. Tumor Necrosis Factor: Methods and Protocols, edited by Angelo Corti and Pietro Ghezzi, 2004 97. Molecular Diagnosis of Cancer: Methods and Protocols, Second Edition, edited by Joseph E. Roulston and John M. S. Bartlett, 2004 96. Hepatitis B and D Protocols: Volume 2, Immunology, Model Systems, and Clinical Studies, edited by Robert K. Hamatake and Johnson Y. N. Lau, 2004 95. Hepatitis B and D Protocols: Volume 1, Detection, Genotypes, and Characterization, edited by Robert K. Hamatake and Johnson Y. N. Lau, 2004 94. Molecular Diagnosis of Infectious Diseases, Second Edition, edited by Jochen Decker and Udo Reischl, 2004 93. Anticoagulants, Antiplatelets, and Thrombolytics, edited by Shaker A. Mousa, 2004 92. Molecular Diagnosis of Genetic Diseases, Second Edition, edited by Rob Elles and Roger Mountford, 2004 91. Pediatric Hematology: Methods and Protocols, edited by Nicholas J. Goulden and Colin G. Steward, 2003 90. Suicide Gene Therapy: Methods and Reviews, edited by Caroline J. Springer, 2004 89. The Blood–Brain Barrier: Biology and Research Protocols, edited by Sukriti Nag, 2003 88. Cancer Cell Culture: Methods and Protocols, edited by Simon P. Langdon, 2003 87. Vaccine Protocols, Second Edition, edited by Andrew Robinson, Michael J. Hudson, and Martin P. Cranage, 2003 86. Renal Disease: Techniques and Protocols, edited by Michael S. Goligorsky, 2003 85. Novel Anticancer Drug Protocols, edited by John K. Buolamwini and Alex A. Adjei, 2003

M E T H O D S I N M O L E C U L A R M E D I C I N E™

Human Papillomaviruses Methods and Protocols

Edited by

Clare Davy John Doorbar Division of Virology, The National Institute for Medical Research Mill Hill, London, UK

© 2005 Humana Press Inc. 999 Riverview Drive, Suite 208 Totowa, New Jersey 07512 www.humanapress.com All rights reserved. No part of this book may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording, or otherwise without written permission from the Publisher. Methods in Molecular Medicine™ is a trademark of The Humana Press Inc. All papers, comments, opinions, conclusions, or recommendations are those of the author(s), and do not necessarily reflect the views of the publisher. This publication is printed on acid-free paper. ∞ ANSI Z39.48-1984 (American Standards Institute) Permanence of Paper for Printed Library Materials. Production Editor: Tracy Catanese Cover design by Patricia F. Cleary Cover illustration: The image shows a section through a papilloma (wart) caused by Human Papillomavirus Type 1 stained to reveal the viral E4 protein (green) and the cellular protein CyclinB (red), which regulates proper entry into mitosis. Cyclin B is readily detected in cells expressing E4. In HPV1 warts, E4 expression begins as soon as the infected cell leaves the basal layer. Nuclei are counter-stained with DAPI (blue), revealing the location of the basal cells surrounding the papillae. Artwork provided by John Doorbar. Photocopy Authorization Policy: Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, is granted by Humana Press Inc., provided that the base fee of US $30.00 per copy is paid directly to the Copyright Clearance Center at 222 Rosewood Drive, Danvers, MA 01923. For those organizations that have been granted a photocopy license from the CCC, a separate system of payment has been arranged and is acceptable to Humana Press Inc. The fee code for users of the Transactional Reporting Service is: [1-58829373-4/05 $30.00]. Printed in the United States of America. 10 9 8 7 6 5 4 3 2 1 e-ISBN 1-59259-982-6 ISSN 1543-1894 Library of Congress Cataloging in Publication Data Human papillomaviruses : methods and protocols / Edited by Clare Davy, John Doorbar. p. ; cm. -- (Methods in molecular medicine ; 119) Includes bibliographical references and index. ISBN 1-58829-373-4 (alk. paper) 1. Papillomavirus diseases--Laboratory manuals. 2. Papilloma-viruses--Laboratory manuals. Clare. II. Doorbar, John. III. Series. [DNLM: 1. Papillomavirus, Human--genetics--Laboratory Manuals. 2. Papillomavirus, Human--pathogenicity--Laboratory Manuals. 3. Cervix Neoplasms--virology--Laboratory Manuals. W1 ME9616JM v.119QW 25 / H918 2005] QR201.P26H86 2005 616.9'11--dc22

I. Davy,

2005006265

Preface Papillomaviruses attract the attention of virologists and doctors alike, perhaps most importantly because human papillomaviruses (HPV) are the etiological agent of cervical cancer, the second most commonly found female cancer worldwide. Historically, research into HPVs has been hampered by the fact that, unlike many other viruses, HPVs show both species and tissue specificity. To overcome this problem, specialized techniques, such as the use of organotypic “raft” cultures to study the HPV lifecycle in human tissue, have been developed. This approach complements the traditional method of histochemistry used on clinical samples and the multitude of molecular methods available for analyzing individual viral protein functions. Despite recent progress on vaccine development, it seems likely that, for the foreseeable future, HPV will remain an important human pathogen. Indeed, fundamental questions regarding both the virus lifecycle and cancer progression remain. It is our hope that Human Papillomaviruses will be a useful tool in helping to find the answers. We have aimed to provide a collection of protocols that will be a useful resource for both basic scientists and clinicians working in the field of papillomavirus research. Although it is impossible to cover all aspects of papillomavirus research, Human Papillomaviruses aims to incorporate a broad range of protocols. Some protocols are already well established, whereas others have been developed only in the last few years. The major themes of this book include: the detection and typing of papillomavirus infections, the study of the papillomavirus life cycle, and the production and functional analysis of papillomavirus proteins. This is achieved using a wide variety of techniques, from PCR to propagation of HPV in vitro. Each chapter has been compiled by experts in the field who are well aware of the pitfalls of their experiments. With this in mind, emphasis has been placed on providing methods that go beyond the details provided in typical journal articles. The protocols are intended to be immediately understandable to a novice in the field, and potential problems are highlighted before they can occur. Of course books like Human Papillomaviruses require input from a large number of people and we are indebted to all the authors for giving up their time to produce such excellent contributions. We would like to thank them for their tolerance of our editorial interventions and our persistent pestering for corrections and signatures. Valuable assistance in proofreading was provided by

v

vi

Preface

Mahmood Ayub and other members of the Doorbar lab. We are also very grateful to John Walker at the University of Hertfordshire, and the editorial staff at Humana Press, for all the advice and assistance they have provided. Clare Davy John Doorbar

Contents Preface .............................................................................................................. v Contributors .....................................................................................................xi 1 Identification of New Papillomavirus Types Ethel-Michele de Villiers, Corinna Whitley, and Karin Gunst ............. 1 2 Identification of HPV Variants John Cason, Jon Bible, and Christine Mant ........................................ 15 3 Histochemical Analysis of Cutaneous HPV-Associated Lesions Kiyofumi Egawa ................................................................................... 27 4 Histological Analysis of Cervical Intraepithelial Neoplasia Michael Babawale, Rashmi Seth, Adam Christian, Wessam Al-Utayem, Ragini Narula, and David Jenkins ................. 41 5 Detection of Papillomavirus Proteins and DNA in Paraffin-Embedded Tissue Sections Woei Ling Peh and John Doorbar ....................................................... 49 6 Detection and Quantitation of HPV Gene Expression Using Real-Time PCR Rashmi Seth, John Rippin, Li Guo, and David Jenkins ....................... 61 7 Analysis of p16INK4a and Integrated HPV Genomes as Progression Markers Svetlana Vinokurova, Nicolas Wentzensen, and Magnus von Knebel Doeberitz ................................................ 73 8 Use of Biomarkers in the Evaluation of CIN Grade and Progression of Early CIN Jan P. A. Baak and Arnold-Jan Kruse .................................................. 85 9 HPV DNA Detection and Typing in Cervical Scrapes Peter J. F. Snijders, Adriaan J. C. van den Brule, Marcel V. Jacobs, René P. Pol, and Chris J. L. M. Meijer .......................................... 101 10 HPV DNA Detection and Typing in Inapparent Cutaneous Infections and Premalignant Lesions Maurits de Koning, Linda Struijk, Mariet Feltkamp, and Jan ter Schegget ..................................................................... 115 11 Establishing HPV-Containing Keratinocyte Cell Lines From Tissue Biopsies Margaret Anne Stanley ..................................................................... 129

vii

viii

Contents

12 Using an Immortalized Cell Line to Study the HPV Life Cycle in Organotypic “Raft” Cultures Paul F. Lambert, Michelle A. Ozbun, Asha Collins, Sigrid Holmgren, Denis Lee, and Tomomi Nakahara ................... 13 Differentiation of HPV-Containing Cells Using Organotypic “Raft” Culture or Methylcellulose Regina Wilson and Laimonis A. Laimins ........................................... 14 Propagation of Infectious, High-Risk HPV in Organotypic “Raft” Culture Margaret E. McLaughlin-Drubin and Craig Meyers.......................... 15 Retrovirus-Mediated Gene Transfer to Analyze HPV Gene Regulation and Protein Functions in Organotypic “Raft” Cultures N. Sanjib Banerjee, Louise T. Chow, and Thomas R. Broker ........... 16 The HPV Xenograft Severe Combined Immunodeficiency Mouse Model William Bonnez ................................................................................ 17 The Cottontail Rabbit Papillomavirus Model of High-Risk HPV-Induced Disease Janet L. Brandsma ............................................................................. 18 Studying the HPV Life Cycle in 3A Trophoblasts and Resulting Pathophysiology Yong Liu, Hong You, and Paul L. Hermonat .................................... 19 Replication and Encapsidation of Papillomaviruses in Saccharomyces cerevisiae Peter C. Angeletti .............................................................................. 20 Analysis of the Regulation of Viral Transcription Bernd Gloss, Mina Kalantari, and Hans-Ulrich Bernard .................. 21 Analysis of HPV Transcription by RPA Jason M. Bodily and Craig Meyers .................................................... 22 Analysis of Regulatory Motifs Within HPV Transcripts Sarah A. Cumming and Sheila V. Graham ........................................ 23 Detection of HPV Transcripts by Nested RT-PCR Christine Mant, Barbara Kell, and John Cason ................................. 24 Analysis of HPV DNA Replication Using Transient Transfection and Cell-Free Assays Biing Yuan Lin, Thomas R. Broker, and Louise T. Chow ..................

141

157

171

187

203

217

237

247 261 279 291 317

331

Contents

ix

25 Detection and Quantitation of HPV DNA Replication by Southern Blotting and Real-Time PCR Iain M. Morgan and Ewan R. Taylor ................................................. 26 Analysis of E7/Rb Associations Sandra Caldeira, Wen Dong, and Massimo Tommasino .................. 27 Transformation Assays for HPV Oncoproteins Paola Massimi and Lawrence Banks ................................................. 28 Analysis of Adeno-Associated Virus and HPV Interaction Paul L. Hermonat, Hong You, C. Maurizio Chiriva-Internati, and Yong Liu ................................................................................. 29 In Vitro Assays of Substrate Degradation Induced by High-Risk HPV E6 Oncoproteins Miranda Thomas and Lawrence Banks ............................................. 30 Measuring the Induction or Inhibition of Apoptosis by HPV Proteins Anna M. Kowalczyk, Geraldine E. Roeder, Katie Green, David J. Stephens, and Kevin Gaston ........................................... 31 Codon Optimization of Papillomavirus Genes Martin Müller .................................................................................... 32 Generation of HPV Pseudovirions Using Transfection and Their Use in Neutralization Assays Christopher B. Buck, Diana V. Pastrana, Douglas R. Lowy, and John T. Schiller ...................................................................... 33 Generation and Application of HPV Pseudovirions Using Vaccinia Virus Martin Sapp and Hans-Christoph Selinka ......................................... Index ............................................................................................................

ix

349 363 381

397

411

419 433

445

463 483

Contributors WESSAM AL-UTAYEM, MD, MS • Division of Histopathology, University of Nottingham Medical School, Queens Medical Centre, Nottingham, UK PETER C. ANGELETTI, PhD • Nebraska Center for Virology, School of Biological Sciences, University of Nebraska-Lincoln, Lincoln, NE JAN P. A. BAAK, MD, PhD, FRCPath, FIAC (HON), Dr. HC (ANTWERP) • Department of Pathology, Stavanger University Hospital, Stavanger, and The Gade Institute, University of Bergen, Bergen, Norway, and Free University, Amsterdam, The Netherlands MICHAEL BABAWALE, MD, PhD • Division of Histopathology, University of Nottingham Medical School, Queens Medical Centre, Nottingham, UK N. SANJIB BANERJEE, PhD • Department of Biochemistry and Molecular Genetics, University of Alabama at Birmingham, Birmingham, AL LAWRENCE BANKS, PhD • Tumour Virology, International Centre for Genetic Engineering and Biotechnology, Trieste, Italy HANS-ULRICH BERNARD, PhD • Department of Molecular Biology and Biochemistry, University of California, Irvine, CA JON BIBLE • Department of Immunobiology, Guy's, King's College and St. Thomas' Hospitals School of Medicine, King's College London, Guy's Campus JASON M. BODILY, MS • The Department of Microbiology and Immunology, The Pennsylvania State University College of Medicine, Hershey, PA WILLIAM BONNEZ, MD • Infectious Diseases Unit, Department of Medicine, University of Rochester School of Medicine and Dentistry, Rochester, NY JANET L. BRANDSMA, PhD • Department of Comparative Medicine, Yale University School of Medicine, New Haven, CT THOMAS R. BROKER, PhD • Department of Biochemistry and Molecular Genetics, University of Alabama at Birmingham, Birmingham, AL CHRISTOPHER B. BUCK, PhD • Laboratory of Cellular Oncology, Center for Cancer Research, National Cancer Institute, Bethesda, MD SANDRA CALDEIRA, PhD • Instituto de Medicina Molecular, Faculdade de Medicina, Universidade de Lisboa, Lisbon, Portugal JOHN CASON, PhD • The Department of Infectious Diseases, Guy’s, King’s College and St Thomas’ Medical and Dental Schools, King’s College, London, UK

xi

xii

Contributors

C. MAURIZIO CHIRIVA-INTERNATI, PhD • Department of Microbiology and Immunology, Texas Tech University, Lubbock, TX LOUISE T. CHOW, PhD • Department of Biochemistry and Molecular Genetics, University of Alabama at Birmingham, Birmingham, AL ADAM CHRISTIAN, BSc, MBBS • Division of Histopathology, University of Nottingham Medical School, Queens Medical Centre, Nottingham, UK ASHA COLLINS • McArdle Laboratory for Cancer Research, University of Wisconsin Medical School, Madison, WI SARAH A. CUMMING, PhD • Institute of Biomedical and Life Sciences, Division of Virology, University of Glasgow, Glasgow Scotland, UK MAURITS DE KONING, MSc • Department of Medical Microbiology, Center of Infectious Diseases, Leiden University Medical Center, Leiden, The Netherlands and Delft Diagnostic Laboratory, Delft, The Netherlands ETHEL-MICHELE DE VILLIERS, PhD • Division for the Characterization of Tumourviruses, Deutsches Krebsforschungszentrum, Heidelberg, Germany WEN DONG, BSc • International Agency for Research on Cancer, World Health Organization, Lyon, France JOHN DOORBAR, PhD • Division of Virology, The National Institute for Medical Research, Mill Hill, London, UK KIYOFUMI EGAWA, MD, PhD • Department of Dermatology, Kumamoto University School of Medicine, Kumamoto, Japan MARIET FELTKAMP, MD, PhD • Department of Medical Microbiology, Center of Infectious Diseases, Leiden University Medical Center, Leiden, The Netherlands KEVIN GASTON, PhD • Department of Biochemistry, School of Medical Sciences, University of Bristol, Bristol, UK BERND GLOSS, PhD • Neurotransgenic Laboratory, Duke University Medical Center, Durham, NC SHEILA V. GRAHAM, PhD • Institute of Biomedical and Life Sciences, Division of Virology, University of Glasgow, Glasgow, Scotland, UK KATIE GREEN, BSc • Department of Biochemistry, School of Medical Sciences, University of Bristol, Bristol, UK KARIN GUNST • Division for the Characterization of Tumourviruses, Deutsches Krebsforschungszentrum, Heidelberg, Germany LI GUO • Department of Histopathology, Division of Molecular Medicine, Queens Medical Centre, University Hospital, University of Nottingham, UK PAUL L. HERMONAT, PhD • Departments of Internal Medicine and Obstetrics and Gynecology, University of Arkansas for Medical Sciences, Little Rock, AR

Contributors

xiii

SIGRID HOLMGREN • McArdle Laboratory for Cancer Research, University of Wisconsin Medical School, Madison, WI MARCEL V. JACOBS, PhD • Section of Molecular Pathology, Department of Pathology, VU University Medical Center, The Netherlands DAVID JENKINS, MD, FRCPath • Department of Histopathology, Division of Molecular Medicine, Queens Medical Centre, University Hospital, University of Nottingham, UK MINA KALANTARI, PhD • Department of Molecular Biology and Biochemistry, University of California, Irvine, CA BARBARA KELL, PhD • The Department of Infectious Diseases, Guy’s, King’s College and St Thomas’ Medical and Dental Schools, King’s College, London, UK ANNA M. KOWALCZYK, BSc • Department of Biochemistry, School of Medical Sciences,University of Bristol, Bristol, UK ARNOLD-JAN KRUSE, MD, PhD • Department of Pathology, Stavanger University Hospital, Stavanger, Norway LAIMONIS A. LAIMINS, PhD • Department of Microbiology-Immunology, Feinberg School of Medicine, Northwestern University, Chicago, IL PAUL F. LAMBERT, PhD • McArdle Laboratory for Cancer Research, University of Wisconsin Medical School, Madison, WI DENIS LEE • McArdle Laboratory for Cancer Research, University of Wisconsin Medical School, Madison, WI BIING YUAN LIN, MS • Department of Biochemistry and Molecular Genetics, University of Alabama at Birmingham, Birmingham, AL YONG LIU, MD, PhD • Departments of Obstetrics and Gynecology, University of Arkansas for Medical Sciences, Little Rock, AR DOUGLAS R. LOWY, MD • Laboratory of Cellular Oncology, Center for Cancer Research, National Cancer Institute, Bethesda, MD CHRISTINE MANT, PhD • The Department of Infectious Diseases, Guy’s, King’s College and St Thomas’ Medical and Dental Schools, King’s College, London, UK PAOLA MASSIMI, PhD • International Centre for Genetic Engineering and Biotechnology, Trieste, Italy MARGARET E. MCLAUGHLIN-DRUBIN, PhD • The Department of Microbiology and Immunology, The Pennsylvania State University College of Medicine, Hershey, PA CHRIS J.L.M. MEIJER, MD, PhD • VSection Molecular Pathology, Department of Pathology, VU University Medical Center, The Netherlands CRAIG MEYERS, PhD • The Department of Microbiology and Immunology, The Pennsylvania State University College of Medicine, Hershey, PA

xiv

Contributors

IAIN M. MORGAN, PhD • Institute of Comparative Medicine (Pathology), University of Glasgow, Glasgow, Scotland MARTIN MÜLLER, PhD • Programme Infection and Cancer, German Cancer Research Center, Heidelberg, Germany TOMOMI NAKAHARA, PhD • McArdle Laboratory for Cancer Research, University of Wisconsin Medical School, Madison, WI RAGINI NARULA, BSc, MBBS • Division of Histopathology, University of Nottingham Medical School, Queens Medical Centre Nottingham, UK MICHELLE A. OZBUN, PhD • Department of Molecular Genetics and Microbiology, University of New Mexico School of Medicine, Albuquerque, New Mexico DIANA V. PASTRANA, PhD • Laboratory of Cellular Oncology, Center for Cancer Research, National Cancer Institute, Bethesda, MD WOEI LING PEH, PhD • Division of Virology, The National Institute for Medical Research, Mill Hill, London, UK RENÉ P. POL, BSc • Section Molecular Pathology, Department of Pathology, VU University Medical Center, The Netherlands JOHN RIPPIN • Department of Histopathology, Division of Molecular Medicine, Queens Medical Centre, University Hospital, University of Nottingham, UK GERALDINE E. ROEDER, LLB, PhD • Department of Biochemistry, School of Medical Sciences,University of Bristol, Bristol, UK MARTIN SAPP, PhD • Institute for Medical Microbiology and Hygiene, University of Mainz, Mainz, Germany JAN TER SCHEGGET, MD, PhD • Department of Medical Microbiology, Center of Infectious Diseases, Leiden University Medical Center, Leiden, and Delft Diagnostic Laboratory, Delft, The Netherlands JOHN T. SCHILLER, PhD • Laboratory of Cellular Oncology, Center for Cancer Research, National Cancer Institute, Bethesda, MD HANS-CHRISTOPH SELINKA, PhD • Institute for Medical Microbiology and Hygiene, University of Mainz, Mainz, Germany RASHMI SETH, PhD • Department of Histopathology, Division of Molecular Medicine, Queens Medical Centre, University Hospital, University of Nottingham, UK PETER J.F. SNIJDERS, PhD • Section Molecular Pathology, Department of Pathology, VU University Medical Center, The Netherlands MARGARET ANNE STANLEY, MB, PhD • Department of Pathology, University of Cambridge, Cambridge, UK DAVID J. STEPHENS, BSc, PhD • Department of Biochemistry, School of Medical Sciences,University of Bristol, Bristol, UK

Contributors

xv

LINDA STRUIJK, PhD • Department of Medical Microbiology, Center of Infectious Diseases, Leiden University Medical Center, Leiden, The Netherlands EWAN R. TAYLOR, PhD • Department of Molecular Biology, The Scripps Research Institute, La Jolla, CA MIRANDA THOMAS, PhD • Tumour Virology, International Centre for Genetic Engineering and Biotechnology, Trieste, Italy MASSIMO TOMMASINO, PhD • International Agency for Research on Cancer, World Health Organization, Lyon, France ADRIAAN J. C. VAN DEN BRULE, PhD • Laboratory for Pathology, PAMM Institute, Eindhoven, The Netherlands SVETLANA VINOKUROVA, PhD • Institute of Molecular Pathology, University of Heidelberg, Heidelberg, Germany MAGNUS VON KNEBEL DOEBERITZ, MD, PhD • Institute of Molecular Pathology, University of Heidelberg, Heidelberg, Germany NICOLAS WENTZENSEN, MD • Institute of Molecular Pathology, University of Heidelberg, Heidelberg, Germany HONG YOU, MD, PhD • Departments of Internal Medicine and Obstetrics and Gynecology, University of Arkansas for Medical Sciences, Little Rock, AR CORINNA WHITLEY • Division for the Characterization of Tumourviruses, Deutsches Krebsforschungszentrum, Heidelberg, Germany REGINA WILSON, BS • Department of Microbiology-Immunology, Feinberg School of Medicine, Northwestern University, Chicago, IL

Identification of New Papillomavirus Types

1

1 Identification of New Papillomavirus Types Ethel-Michele de Villiers, Corinna Whitley, and Karin Gunst Summary The identification of papillomavirus DNA sequences in tissue samples using polymerase chain reaction (PCR) amplification, has led to the association of these infections to a multiplicity of clinical manifestations. The cloning and sequencing of PCR-amplified products has, to date, resulted in the identification of more than 300 putative “new” papillomavirus types. The methods used to identify these unknown papillomavirus sequences are described here. The CP, FAP, and GP primers are used for PCR amplification, followed by cloning and sequencing of the amplicons. Sequence comparisons and the interpretation of DNA sequence identities are discussed. Details of defining a new papillomavirus type and of the recently approved taxonomic classification system for the Papillomaviridae are given.

1. Introduction The identification and isolation of new papillomavirus types have been an ongoing process spanning almost three decades. The lack of cell-culture systems for the in vitro propagation of these viruses hampered investigations considerably. The first human papillomavirus (HPV) DNA was isolated in the late 1970s, at a time when the use of restriction enzymes, cloning, and sequencing was a novelty. Initially, papillomaviruses were purified by caesium-chloride gradient centrifugation of minced warts, followed subsequently by DNA extraction from the fractions containing viral particles. The application of the Southern blot hybridization technique enabled the identification of closely related papillomavirus types. Restriction enzyme-digested DNA samples from various tumors were separated by gel electrophoresis and transferred onto nitrocellulose membranes. A radiolabeled DNA probe of a known papillomavirus type was hybridized to the membrane at a given temperature below the DNA melting point. Identical or related sequences were detected by varying these From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

1

2

de Villiers, Whitley, and Gunst

hybridization conditions. In addition, comparative analyses of restriction fragment lengths induced by individual or combinations of restriction enzymes allowed for the identification of differences in the genomes of the respective isolates. The small group of scientists involved in papillomavirus research at that stage agreed to accept an isolate as a papillomavirus only if the complete genome had been cloned and made available. The full-length viral genomes were isolated from tissues either by ethidiumbromide density gradient centrifugation from tissues containing many copies of papillomavirus particles, or by plaque hybridization of bacteriophage libraries constructed from digested total cellular DNA. Gradient-purified viral DNA was cloned directly into a bacterial plasmid, whereas virus DNA integrated into the bacteriophage genome was subsequently subcloned into a plasmid. The degree of DNA homology between the full-length genomes of related papillomavirus types was determined by liquid hybridization and, in some instances, by analyzing the heteroduplex formation of the genomes by electron microscopy. A new papillomavirus type was then defined as sharing less than 50% homology in liquid hybridization to the closest related known papillomavirus type. The advent of the sequencing technology facilitated the classification of papillomaviruses. Individual genes within the viral genome were identified, and the genome organization characteristic for all papillomaviruses was determined. The sequence homology between individual genes from a number of papillomavirus types could now be compared, and conserved regions within the genome identified. This development led to a modification of the definition of a new papillomavirus type. The members of the papillomavirus research community this time agreed on defining a new papillomavirus type as follows: a complete papillomavirus genome sharing less than 90% DNA homology in each of the E6, E7, and L1 open reading frames (ORF) to the respective ORFs of the closest related papillomavirus type. The L1 gene is the most conserved ORF, and its protein is the main viral capsid protein. The choice of E6 and E7 was based on their functional importance. During the beginning of the 1990s, the majority of papillomavirus genomes were sequenced by Hajo Delius (1). With all the sequence data available, it became evident that the decision to include the E6 and E7 sequences could not be upheld. The HPV 77 E7 shares 97% DNA sequence homology to the HPV 29 E7 (2). It was therefore decided to revert to comparing only the L1 ORF when defining a new papillomavirus type. This has remained the definition of a papillomavirus type to the present time (reviewed in ref. 3). The possibility of detecting papillomavirus DNA sequences in tissue by polymerase chain reaction (PCR) amplification has led to a very rapid increase in the number of putative new papillomavirus types being identified. The enormous interest in the papillomavirus types associated with genital lesions and


3

the potential use for diagnostic purposes, led to the development of the MY09/ MY11 primers (4) amplifying a region of the L1 ORF of a broad spectrum of the types belonging to the genus α-papillomavirus. A large number of putative new HPV types were also identified by sequencing the amplicons (5,6). The GP5+/GP6+ primers amplify a shorter region within the same region of the L1 ORF (7). Several other PCR primer systems amplifying regions of the E6, E7, E1, or L1 ORFs were reported, but used only in single laboratories. The detection of papillomavirus sequences in cutaneous lesions requires the use of PCR primers that amplify a broader spectrum of papillomavirus types. HPV types in the genera β-papillomavirus, γ-papillomavirus, µ-papillomavirus, as well as a number of HPV types in the genus α-papillomavirus were originally isolated from cutaneous lesions. Emerging data on the detection of papillomavirus sequences in tumor and normal tissue from a variety of organs indicate the existence of an extremely large group of papillomavirus types. Unfortunately, very little information is available on the pathogenic mechanisms of the majority of individual papillomavirus types. Earlier notions failed to modify the definition of a type in order to combine types sharing high DNA sequence homology, due to the divergence in their biological behavior. An example can be found in certain cutaneous HPV types, which each react completely differently upon ultraviolet irradiation despite their genomes sharing a very high DNA homology (8). There is a clear need to continue the process of identifying unknown papillomaviruses. Phylogenetic analysis (3) is suggestive of the possible existence of larger numbers of papillomavirus types in certain genera, which, to date, have not been studied extensively. On the other hand, histological data of certain types of tumors are indicative of the possible involvement of an infection with papillomavirus. Combining these data, as well as extending ongoing studies, may lead us to the identification of additional factors in the etiology of disease. The following describes a useful approach to identify unknown papillomaviruses, independent of the organ origin of the cellular DNA. (Read notes carefully prior to preparing for or starting with any experiments.) 2. Materials 2.1. DNA Extraction 1. Phenol (pH stabilized with 10 mM Tris, 1 mM ethylenediamine tetraacetic acid [EDTA], pH 8.0). 2. Proteinase K (20 mg/mL). 3. PK buffer (2X): 0.2 M Tris-HCl (pH 7.5), 25 mM EDTA, 0.3 M NaCl, 2% sodium dodecyl sulfate (SDS). Do not autoclave this solution, but sterilize through filtration. 4. Chloroform:isoamylalcohol (CIA) at 24:1.

4

de Villiers, Whitley, and Gunst 5. 6. 7. 8. 9.

Ethanol, absolute, 70%, 80%, and 90%. Xylol. 1X TE : 10 mM Tris-HCl,1 mM EDTA (pH 7.4). 3 M Na acetate (pH 5.4). Orbital mixer.

2.2. Polymerase Chain Reaction 1. dNTP set: mixture containing 10 mM each dNTP. Aliquot and store at –20°C until use. 2. Primers: aliquot at a dilution of 10 pmol/µL and store at –20°C until use. For GP amplification (7)—modified: L1 ORF: amplicon size 140–150 bp GP 5+: 5'–ttggatccT TTG TTA CTG TGG TAG ATA CTA C–3' GP 6+: 5'–ttggatccG AAA AAT AAA CTG TAA ATC ATA TTC–3' For CP amplification (9)—modified: L1 ORF: amplicon size 480–500 bp, and for the nested reaction, 370–390 bp CP 65: 5'–ttggatccC ARG GTC AYA AYA ATG GYA T–3' CP 70: 5'–ttggatccA AYT TTC GTC CYA RAG RAW ATT GRT C–3' CP 66: 5'–ttggatccA ATC ARM TGT TTR TTA CWG T–3' CP 69: 5'–ttggatccG WTA GAT CWA CAT YCC ARA A–3' For FAP amplification (10)—modified: L1 ORF: amplicon size approx 480 nucleotides FAP59 (forward): 5'–ttg ga tccT AAC WGT IGG ICA YCC WTA TT–3' FAP64 (backward): 5'–ttg gat ccC CWA TAT CWV HCA TIT CIC CAT C–3' The BamHI restriction site (ggatcc) and the tt overhang were incorporated into all primers to facilitate cloning. Y: C or T; R: A or G; W: A or T; M: A or C; 3. Taq polymerase (Ampli Taq Gold; Perkin Elmer). The PCR buffer (10X) and MgCl2 (25 mM) solution are included. Aliquot these and store at –20°C until use. MgCl2 (25 mM) is added to a final concentration of 2 mM for the CP and GP primers and 3.5 mM for the FAP primers. 4. Sample DNA (50–100 ng in a total volume of 10 µL). 5. Size marker: DNA ladder mix (100–10,000 bp) 6. Loading buffer (xylene cyanol: 0.5% xylene cyanol, 50% glycerol). 7. EP buffer: 40 mM Tris-HCl, 5 mM Na acetate, 1 mM EDTA (pH 7.8). 8. Agarose gels (1.5% and 2%) in 1X EP buffer. 9. Ethidium bromide (concentration 0.1%). 10. PCR thermocycler: Multicycler PTC200 (Biozym Diagnostik).

2.3. Cloning of the Amplicons 1. High Pure PCR Product Purification Kit, Roche (Cat. No. 1732676). 2. Gel extraction kit, e.g., JETQUICK (Genomed Cat. No. 420050). 3. TA Cloning Kit complete with TOP10F' competent cells (Invitrogen).


5

4. Luria-Bertani (LB) broth (4 g tryptone, 4 g NaCl, 2 g yeast extract) containing ampicillin (end concentration 100 µg/mL). 5. LB agar plates (LB broth and 6 g Bacto-agar) containing 100 µg/mL ampicillin, 40 mg/mL isopropyl-β-D-thiogalactopyranoside (IPTG) and 40 µg/mL X-gal (5-bromo-4-chloro-3-indolyl-β-galactoside). Ampicillin, IPTG, and X-gal should be added to the solutions after sterilization. Take care to cool solutions partially prior to addition. 6. Plasmid miniprep kit, e.g., JETQUICK (Genomed). 7. Restriction enzyme BamH1.

2.4. Sequencing and Sequence Analysis 1. Automated sequencer. 2. Sequence analysis package software.

3. Methods 3.1. DNA Extraction (see Notes 1–3) If fresh biopsies are used, DNA is extracted either directly or after storage at –70°C until use. 1. Cut tissue into very small pieces and place into 1.5-mL tube. 2. Add absolute ethanol and leave overnight at room temperature. 3. Remove ethanol and lyophilize pellet (cover open tube with parafilm and pierce several times with needle). 4. Add 250 µL PK buffer, 250 µL double-distilled water, and 10 µL proteinase K (20 mg/mL) to tissue. 5. Close tube, seal with parafilm, and rotate the tube for at least 5 h or overnight on an orbital mixer at 37°C. 6. Add 500 µL phenol (saturated and stabilized at pH8 with 1X TE) and mix well by rotation for at least 10 min. 7. Centrifuge for 10 min at 13,000g. 8. Aspirate supernatant into a new 2-mL tube and add 250 µL phenol and 250 µL CIA. 9. Repeat rotation and centrifugation. 10. Repeat step 8 and transfer supernatant to a new 2-mL tube containing 500 µL CIA. 11. Repeat step 9. 12. Aspirate supernatant into new 1.5-mL tube after rotation and centrifugation— take great care not to aspirate CIA when removing supernatant. 13. Add 1/10 volume Na acetate (pH 5.4) and 2 volumes absolute ethanol. 14. Cool at –20°C for at least 1 h (preferably overnight). 15. Centrifuge tube in pre-cooled centrifuge for 30 min at 13,000g. 16. Discard supernatant and add 70% ice-cold ethanol to wash pellet by centrifugation for 15 min at 13,000g. 17. Remove supernatant carefully and lyophylize pellet. 18. Dissolve pellet according to size in 1X TE (pH 8.0). 19. Measure the DNA concentration.

6


3.2. Removal of Paraffin Prior to Extracting the DNA in the Case of Paraffin Embedded Tissue (see Notes 2–5) Paraffin is removed as follows, depending on the number of sections and the thickness of single sections. For three to four sections of 3–5 µm thickness in 1.5-mL tube: 1. 2. 3. 4. 5. 6. 7. 8. 9.

Add 200 µL xylol mix thoroughly by rotation on a mixer for at least 10 min. Remove supernatant after centrifugation for 10 min at 13,000g. Repeat steps 1 and 2 twice. Add 200 µL absolute ethanol and repeat mixing, centrifugation, and aspiration of supernatant as above. Repeat as above, except for replacing the supernatant with 90%, 80%, and 70% ethanol in successive steps. Close tube with parafilm and pierce with needle. Lyophilize. Store tightly closed at 4°C until further use, or Proceed to steps 4–19 as described under Subheading 3.1.

For three sections of 25–30 µm thickness: 1. 2. 3. 4. 5. 6. 7.

Add 1 mL xylol in 1.5-mL tube. Rotate overnight on orbital mixer. Centrifuge for 10 min at 13,000g. Aspirate and discard supernatant. Add 1 mL absolute ethanol. Repeat steps 2–4. Repeat steps 1–9 as above, except that a volume of 1 mL of the respective solutions is added.

3.3. PCR Amplification (see Notes 1, 2, and 6) 3.3.1. Amplification Using the CP primers 1. The first round of PCR is done using the CP65 and CP70 primers. Prepare the following master mix for the PCR amplification according to the number of samples to be tested: 5 µL 10X PCR buffer, 1 µL of 4 × 10 mM dNTPs, 5 µL each of 10 pmol/µL CP65 and CP70, 0.25 µL 5 U/mL Taq polymerase, and 5 µL 25 mM MgCl 2 in a total volume of 40.25 µL. 2. Add the sample DNA to the tube containing an aliquot of the master mix. Denature at 94°C for 9 min and amplify through 40 cycles of 60 s at 94°C, annealing for 60 s at 50°C, and elongation for 1 min at 72°C. Add a final elongation step of 5 min at 72°C. 3. Run 10 µL per amplified product with 2 µL loading buffer on a 1.5% agarose gel and visualize bands by staining with ethidium bromide (expected amplicon size CP65/CP70 is 452–467 bp) (see Notes 7–9).


7

4. Aspirate 3 µL of the first amplified product for the subsequent nested PCR using the primers CP66 and CP69. The same PCR conditions are used as for CP65/CP70. 5. Repeat step 3 for visualization of the amplicons (expected size CP66/CP69 is 374–389 bp) (see Note 9). 6. If a single band is visible after EtBr staining, purify the total amplification product using the High Pure PCR Product Purification Kit. 7. If more than one band is visible, precipitate the sample (refer to steps 14–16 under Subheading 3.1.). Dissolve the pellet in 10 µL 1X TE (pH 8.0), add loading buffer, and separate on a 1.5% agarose gel; stain and cut each band from gel with a sterile scalpel (see Note 10). 8. Purify the amplicons excised from the agarose gel using the JETQUICK Gel Extraction Kit. 9. The total volume after elution will be 50 µL. Measure the DNA concentration. 10. Ligate the DNA fragment into one of the mentioned vectors (e.g., pCR2.1) using TA Cloning Kit, at ratio of five insert molecules to one vector molecule (follow instructions as described in the kit) with overnight ligation at 14°C. 11. Transform TOP10F' competent Escherichia coli with the ligated DNA. 12. Plate bacteria on LB agar plates containing ampicillin, IPTG, and X-gal, and incubate overnight at 37°C. 13. Cool agar plates at 4°C for 1 h in order to increase the contrasting of blue and white colonies. 14. Select white colonies: pick individual colonies each with a sterile toothpick and transfer the latter into a tube containing 5 mL LB broth with ampicillin (see Notes 11 and 12). 15. Shake bacterial cultures by rotation (180 to 220 rpm) at 37°C overnight. 16. Purify the plasmid DNA from the bacteria using a miniprep kit according to the manufacturer’s instructions, eluting in a final volume of 75 µL. 17. Digest 2 µL of each purified DNA sample with the restriction enzyme BamH1. 18. Separate digested fragments on a 1.5% agarose gel. 19. Visualize the DNA bands by EtBr staining and ultraviolet (UV) light. 20. Select the clones with inserts of the expected size (see Note 13). 21. Sequence an aliquot of at least 6–10 clones per amplification product.

3.3.2. Amplification Using the GP5+/GP6+ Primers 1. Amplify the sample DNA using the GP5+/GP6+ primers in the following reaction: 5 µL 10X PCR buffer, 1 µL of 4 × 10 mM dNTPs, 5 µL each of 10 pmol/µL GP5+ and GP6+, 0.25 µL 5 U/mL Taq polymerase, and 5 µL 25 mM MgCl2 in a total volume of 40.25 µL. 2. Denature at 94°C for 9 min and amplify through 40 cycles of 40 s at 94°C, annealing of 90 s at 40°C, and elongation for 90 s at 72°C. Add a final elongation step of 5 min at 72°C. 3. Repeat step 3 under Subheading 3.3.1. The expected amplicon size is 140 bp to 150 bp. 4. Repeat steps 5–21 under Subheading 3.3.1.

8


3.3.3. Amplification Using the FAP Primers 1. Amplify the sample DNA using the FAP primers and the following master mix: 5 µL 10X PCR buffer, 1 µL of 4 × 10 mM dNTPs, 5 µL each of 10 pmol/µL FAP59 and FAP64, 0.25 µL 5 U/mL Taq polymerase, and 7 µL 25 mM MgCl2 in a total volume of 40.25 µL. 2. Denature at 94°C for 8 min and amplify through 45 cycles of 90 s at 94°C, annealing for 90 s at 50°C, and elongation for 90 s at 72°C. Add a final elongation step of 5 min at 72°C. 3. Repeat step 3 under Subheading 3.3.1. The expected amplicon size is 480 nucleotides (see Note 14). 4. Repeat steps 5–21 under Subheading 3.3.1.

3.3.4. Amplification of DNA From Paraffin-Embedded Tissue to Determine Fragment Size Range If DNA from paraffin-embedded tissue is used for amplification by PCR, the integrity of the cellular DNA has to be determined by PCR amplification of a household gene, e.g., β-actin (11), prior to amplification with papillomavirusspecific primers. The average fragment size of the cellular DNA is determined by using different combinations of the primers described. If the maximum length amplified does not exceed, for example, 100 bp, it is unlikely that the PCR amplification using the CP or FAP primers will succeed (see Note 5).

3.4. Sequence Comparisons With Databanks 1. The software chosen for the analyses of sequences obtained from the cloned amplicon compared to the sequences available in the databanks, depends on what is available in the specific laboratory. The HUSAR Package (12) is very useful (see Note 15). 2. Sequences are initially compared to the viral databanks. If sequence identity to any other virus type is listed first, the query sequence probably represents a cellular sequence and not a papillomavirus sequence. 3. If the identity between the query sequence and a papillomavirus sequence in the databank is 90% or above—and covers almost all of the sequence length (see Note 16)—the query sequence is that papillomavirus type or partial putative papillomavirus type (see Note 17). 4. If the identity between the query sequence and the closest related papillomavirus sequence in the databank is below 90%, but above 80%, the query sequence is defined as related to the next closest papillomavirus type, and therefore probably defines a putative new papillomavirus type. To verify this finding, sequencing of both strands of the clone is necessary to exclude sequencing errors and to determine the exact full-length sequence. 5. If the identity between the query sequence and the closest related papillomavirus sequence is lower than 80%, a comparison to all databanks is necessary to determine whether the query sequence may be of cellular origin. If the same result is


9

obtained as against the viral databank, the translation of the query sequence (across all six reading frames) is necessary to determine whether the query sequence harbors the amino acids conserved among papillomaviruses within the amplified region. If the latter applies, the sequence probably represents a putative new papillomavirus distantly related to the closest related papillomavirus type. If it does not apply, the query sequence does not represent a papillomavirus sequence. 6. In all cases where putative new papillomavirus sequences are identified, subsequent attempts should be made to clone the full-length papillomavirus genome (see Note 18).

4. Notes 1. Probably the most important aspect of applying the polymerase chain reaction is taking precautionary measures to avoid DNA contamination. This applies not only in the laboratory where the testing is being performed, but starts at the point where tissue samples are collected. In all instances, fresh gloves, scalpels, and containers must be used. Freezing multiple samples (tubes) in large liquid-nitrogen containers could lead to crosscontamination, because the tube lids, in most cases, allow for the penetration of the liquid nitrogen into and out of the closed tube. Precautions in the laboratory include repeated cleaning of surfaces, changing of gloves, and separation of handling procedures and instruments into different rooms. An example is to change the laboratory clothing prior to entering the room where the solutions (e.g., master mix in the absence of DNA) are prepared, vs a room where the addition of the sample DNA and amplification steps are performed. Tubes should preferably be centrifuged prior to opening in order to minimize the formation of an aerosol when opening the lid. Even then, the opening should be performed cautiously. Contamination on the outside wall of the tube and lid must be avoided, as well as the contamination of all laboratory instruments. Plugged disposable pipet tips must be used at all times. 2. It is not advisable to handle or amplify more than 5–10 samples at one time. If contamination does occur, it can easily be localized to a specific sample or experiment. In this case, repetition of the experiment starting with the extracted sample DNA is advisable. If this leads to the same result as initially obtained, additional DNA should be extracted again from the original sample and the experiments repeated using new aliquots of all solutions, and so on. 3. Care should be taken when dissolving a lyophilized DNA pellet. Forced pipetting will lead to fragmentation of the DNA sample. It is advisable to leave the buffer on the pellet for some time in order to soften the pellet prior to pipetting. It is advisable to keep the sample DNA in a more concentrated form and to prepare dilutions of aliquots when needed. 4. The sectioning of fixed samples requires the use of a new blade for each individual sample, in combination with an extensive cleaning of the microtome between samples.

10


5. DNA extracted from fixed tissue is often degraded (as a result of inappropriate fixing methods/times). The quality of the extracted DNA should be tested by the amplification of a house-keeping gene (e.g., β-actin [11]) in order to determine the average fragment size of the DNA. In some instances (e.g., lesions producing a high copy number of papillomavirus particles) amplification of larger papillomavirus sequences may be obtained from tissue samples of which the average DNA fragment size is below 200 bp. Apparently intact virus particles are resistant to degradation, and viral genome may be extracted following the described methods. 6. Aliquot all solutions in volumes adequate for one experiment. This requires additional time, but in the end saves more time and effort in instances when contamination does occur. 7. Include negative as well as positive controls with each amplification. Negative control means one sample containing only the master mix without any DNA, as well a sample containing any “neutral” cellular DNA (e.g., placenta DNA). Positive controls will include an HPV type at specific concentration. The following HPV types are suggested as controls: HPV 3 when using GP+ primers, HPV 8 when using CP and FAP primers. Each of these is included at dilutions of 10 and 100 viral copies against a placenta DNA background. If any of the HPV types used as control appears to be present in one of the samples tested, the experiments are repeated using different HPV types as controls. This is done to exclude the possibility of cross-contamination. In this case, HPV 18 is used with the GP+ primers, HPV 23 with the CP primers, and HPV 32 with the FAP primers. 8. Run DNA size markers on both sides of the samples during gel electrophoresis. This allows for a more accurate fragment size determination. 9. When using the CP primers CP65/CP70 for amplification, two nonspecific bands (approx sizes 500 bp and 250 bp) are often seen. The viral DNA fragment size is generally approx 450 bp. If the viral DNA was amplified sufficiently, only one band will be present. In all other cases, or in case of uncertainty, it is advisable to perform the nested PCR reaction using the CP65/CP69 primers. The amplicons generated here are usually of viral origin, but weak bands may indicate low viral copy numbers mixed with cellular sequences. Therefore, cloning and sequencing are necessary steps to verify the origin of the sequences. 10. Do not run more than one sample per gel when running preparative gels. The huge amount of DNA present could easily cross-contaminate in the buffer as well as during handling of the gel. Removal (cutting with a sterile scalpel) of the required DNA band from the agarose gel should be performed under long-wave UV visualization in the shortest period of time possible. Short-wave UV may damage the DNA molecules extensively, leading to fractioning. Place a clean plastic sheet between the UV filter and the gel to avoid contamination of subsequent samples. 11. Initially, 12 bacterial colonies are picked for DNA extraction. If all plasmid preparations contain inserts of the expected size, 6–10 inserts will be sequenced. If less than 6 preparations contain inserts of the expected size, additional bacte-


12.

13.

14.

15.

16.

17.

18.

11

rial colonies will be examined. The possibility of identifying infections with multiple HPV types and/or putative new HPV types (both when either present at low copy numbers or distantly related to known types) increases with the analysis of increasing numbers of colonies/inserts. Aliquots of all bacterial culture suspensions should be preserved for further use or conservation. The same applies for ligation reactions and DNA from plasmid preparations. Larger fragment sizes, expected when using the initial primers, may be obtained after the nested amplification. This results from additional amplification in the second reaction of the initial amplicon by the carrying over of initial primers, and so on, in the sample taken from the first amplification round to initiate the nested reaction. When using the FAP primers, the expected amplicon size is approx 480 bp. Offsized amplicons of approx 200 bp are sometimes observed. Cloning and sequencing of these amplicons indicated the presence of additional binding site(s) for the FAP primers within this stretch of the L1 ORF of certain HPV types. It is therefore necessary to examine such products by cloning and sequencing in order to capture any unidentified (new) HPV sequences. The sequencing reaction is initiated by using the primers located adjacent to the multiple cloning site of the vector. Make sure to delete adjacent vector sequences from the sequence in question prior to comparing insert sequences to other sequences in the databanks. Vector sequences are often included in many of the submitted sequences and may cause confusion in interpreting the results. It is very important that the sequence identity should encompass almost the full length of the cloned DNA. Identity over short stretches could imply the coverage of only the primer sequences, with the rest not being papillomavirus sequences. The PCR amplicon constitutes a partial HPV sequence and may be indicative of a putative new HPV type. The complete genome has to be cloned and sequenced (characterized) in order to be referred to as an HPV type. The complete genomes are then integrated into the taxonomic classification of the Papillomaviridae, which was recently agreed upon by the International Committee on the Taxonomy of Viruses (ICTV) (reviewed in ref. 3). The following definitions for the Papillomaviridae may be used as guide for the interpretation of the cloned sequences: • Genera share less than 60% nucleotide sequence identity in the L1 ORF. • Species share between 60% and 70% nucleotide identity. • Types within species share between 71% and 89% identity. • Nucleotide identity of the L1 ORF of 90% to 98% and more constitutes a subtype, and higher than 98% a variant of a known papillomavirus type. The isolation and cloning of full-length papillomavirus genomes may be tedious if small amounts of sample DNA are available, if low copy numbers are present in the sample, or if multiple types are present within one sample. The direct cloning of the complete genome into vector systems capable of harboring larger fragments is preferable (Bacteriophage lambda, Expand Cloning Kit; Roche, Cat. No. 1940392).

12

de Villiers, Whitley, and Gunst Alternative ways are being used, but these do pose problems. Amplification by long PCR (a number of different kits are available) using outward primers, designed on the initial PCR fragment, may lead to sequence modifications (even though proofreading enzymes are used). It was therefore decided to label HPV types generated by this method as HPVcand . . . (cand = candidate). This method may even lead to the amplification of “hybrid” HPV genomes if more than one HPV type is present in the same sample and these types are closely related types or amplification conditions allow for the nonstringent annealing of the primers. Another method being applied with success is the multiply primed rolling circle amplification using the Phi29 DNA polymerase and random hexamer primers (13). Kits (Amersham, GenomiPhi Cat. No. 25-660-01, and TempliPhi Cat. No. 25-6400-10) are available. Conditions have to be optimized for the amplification of papillomavirus genomes (14). The initial amplification of the cellular DNA using this method will facilitate amplification using the long PCR amplification method (Gunst and de Villiers, unpublished results). The complete genomes generated this way have, to date, not been verified on sequence identity to the papillomavirus sequence present in the original tissue sample.

References 1. Delius, H. and Hofmann, B. (1994) Primer-directed sequencing of human papillomavirus types. Curr. Top. Microbiol. Immunol. 186, 13–31. 2. Delius, H., Saegling, B., Bergmann, K., Shamanin, V., and de Villiers, E-M. (1998) The genomes of three of four novel HPV types, defined by differences of their L1 genes, show high conservation of the E7 gene and the URR. Virology 240, 359–365. 3. de Villiers, E-M., Fauquet, C., Broker, T., Bernard, H-U., and zur Hausen, H. (2004) Classification of papillomaviruses. Virology 324, 17–27. 4. Bauer, H. M., Greer, C. E., Chambers, J. C., et al. (1991) Genital human papillomavirus infection in female university students as determined by a PCRbased method. JAMA 265, 472–477. 5. Bernard, H-U., Chan, S. Y., Manos, M. M., et al. (1994) Identification and assessment of known and novel human papillomaviruses by polymerase chain reaction amplification, restriction fragment length polymorphisms, nucleotide sequence, and phylogenetic algorithms. J. Infect. Dis. 170, 1077–1085. 6. Manos, M. M., Waldman, J., Zhang, T. Y., et al. (1994) Epidemiology and partial nucleotide sequence of four novel genital human papillomaviruses. J. Infect. Dis. 170, 1096–1099. 7. De Rhoda-Husman, A. M., Walboomers, J. M., van den Brule, A. J., Meijer, C. J., and Snijders, P. J. (1995) The use of general primers GP5 and GP6 elongated at their 3' ends with adjacent highly conserved sequences improves human papillomavirus detection by PCR. J. Gen. Virol. 76, 1057–1062. 8. de Villiers, E-M., Ruhland, A., and Sekaric, P. (1999) Human papillomaviruses in non-melanoma skin cancer. Semin. Cancer Biol. 9, 413–422.


13

9. Berkhout, R. J. M., Tieben, L. M., Smits, H. L., Bouwes Bavinck, J. N., Vermeer, B. J., and ter Schegget, J. (1995) Nested PCR approach for detection and typing of Epidermodysplasia Verruciformis-associated human papillomavirus types in cutaneous cancers from renal transplant recipients. J. Clin. Microbiol. 33, 690–695. 10. Forslund, O., Antonsson, A., Nordin, P., Stenquist, B., and Hansson, B. G. (1999) A broad range of human papillomavirus types detected with a general PCR method suitable for analysis of cutaneous tumours and normal skin. J. Gen. Virol. 80, 2437–2443. 11. Greer, C. E., Peterson, S. L., Kiviat, N. B., and Manos, M. M. (1991) PCR amplification from paraffin-embedded tissues. Effects of fixative and fixation time. Am. J. Clin. Pathol. 95, 117–124. 12. Senger, M. T., Flores, K-H., Glatting, P., Hotz-Wagenblatt, A., and Suhau, S. (1998) W2H: WWW interface to the GCG sequence analysis package. Bioinformatics 14, 452–457. 13. Dean, F. B., Nelson, J. R., Giesler, T. L., and Lasken, R. S. (2001) Rapid amplification of plasmid and phage DNA using phi29 DNA polymerase and multiplyprimed rolling circle amplification. Genome Res. 11, 1095–1099. 14. Rector, A., Tachezy, R., and van Ranst, M. (2004) A sequence-independent strategy for detection and cloning of circular DNA virus genomes by using multiply primed rolling-circle amplification. J. Virol. 78, 4993–4998.

Identification of HPV Variants

15

2 Identification of HPV Variants John Cason, Jon Bible, and Christine Mant Summary The vast majority of anogenital carcinomas are caused by high-risk human papillomaviruses (HPVs), and among Western nations HPV-16 is usually the most predominant cancer-associated type. As a DNA virus, HPV type 16 has a relatively stable genome that is believed to have co-evolved with its host over the millennia. Nevertheless, among the “wild” populations of HPV-16 that are circulating, a large number of variants have been identified, and these may have considerably different pathogenic potentials. In this chapter, methods for screening and characterizing HPV-16 sequence variants are described. In particular, we describe methods for the identification of variation within the HPV-16 E5 open reading frame and for the detection of the nt 131 A→G mutation of the E6 ORF, using restriction fragment length polymorphism assays . In addition, we describe approaches for DNA sequencing and analysis. Such methods are likely to be of particular interest to those involved in epidemiological investigations of virus transmission and pathogenicity studies.

1. Introduction Cervical cancer is a major cause of female cancer deaths, with some 450,000 incident cases worldwide (1). It is now clearly established that a subgroup of human papillomaviruses (HPVs) are causally associated with this malignancy and are termed high-risk (HR) HPVs (2). In the United Kingdom—and in most Western countries—HPV types 16 and 18 are most frequently detected HRHPVs in cervical malignancies: in our inner-city location, HR-HPV DNA occurs in about 95% of cervical cancers, and 62% are positive for HPV-16 DNA (3). Because the vast majority of HR-HPV infections do not result in carcinoma (4,5), other factors must be involved in malignant progression. Although co-factors for cervical cancer have been sought, no single convincing co-factor has been identified, and the greatest risk for developing cervical cancer remains persistent infection with a HR-HPV and a high viral load. From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

15

16

Cason, Bible, and Mant

For a long time, there has been interest in the phylogeny of HPV-16 variants (6), and some studies have sought an association between HPV-16 variants and cervical neoplasia (7–11). However, these reports are largely based on crosssectional studies of small patient numbers and a limited clinical spectrum of neoplastic lesions, and rarely include infected women with normal cytology. Additional, more detailed longitudinal studies of the association between HPV-16 variants and disease are required. The identification of HPV-16 variants may also be suited for use in studies of virus transmission for epidemiological purposes and in litigation cases of alleged sexual abuse. In this chapter we describe two restriction fragment length polymorphism (RFLP) assays that can be used to rapidly identify the presence of variations within the E5 and E6 open reading frames (ORFs) of HPV-16 variants in large population studies, and a DNA sequencing strategy to rapidly identify and analyze HPV-16 variants. 2. Materials 2.1. Reference Materials and Clinical Sample Preparation 1. Positive controls: reference isolates of HPV-16 such as pAt-16 (12,13), available from Dr. E. M. DeVilliers, DKFZ, Heidelberg, Germany), and DNA from CaSki or SiHa cells (both are HPV-16 DNA positive; obtainable from the American Type Culture Collection [ATCC] Ltd., Rockwell, MD). 2. Negative controls: an HPV-16 DNA negative cell line (e.g., A431, available from ATCC). 3. Cervical brush smears: typical samples for analysis would include cervical brush smears collected with an Axibrush™ (Colgate Medical Ltd.) from women attending local well-woman centers and gynecological outpatient clinics. 4. Dulbecco’s phosphate-buffered saline (PBS). 5. Proteinase K (PK) solution: 0.45% v/v NP-40, 0.45% v/v Tween-20, 60 g/L PK (Roche Ltd., UK).

2.2. Setting Up PCR Reactions 1. 2. 3. 4. 5.

DNA polymerase (5 U/µL; Promega). 10X polymerase chain reaction (PCR) buffer (Promega). 25 mM MgCl2 (Promega). E5 PCR primers (see Table 1 for sequences and conditions). E6 PCR primers and cycling conditions: The first E6 PCR uses the primers E61A (GAGAACTGCAATGTTTCAGG) and E62A (TGATTA CAGCTGGGTTTCTC: 3) which amplifies a 469-bp fragment of the E6 gene (Fig. 1A). The second primer set consists of E61B (CCAAAAGAGAACTGCAATGT) and E62B (AATTTTAGAATAAAACTTTAAACATT) (Fig. 1B). 6. Molecular biology-grade (MBG) DNase-free water. 7. Premixed dNTPs (Cambio, Ltd.).


17

Table 1 Example of the E5 Polymerase Chain Reaction Conditions Reaction buffer (10X stock) 20 µL

MgCl2 (25 mM)

dNTPs (2.5 mM each stock)

Amount of each primer (2.5 mM)

DNA polymerase (5 U/mL)

dH2O

Sample

20 µL

16 µL

2 µL

1 µL

119 µL

20 µL

Upstream Primer TACAGGATCC TTATGTAATTA AAAAGCGTGC AT

Downstream Primer ATTTAGATC TATATGACA AATCTTGAT ACTGC

Cycling (×40)

(×1) 94°C/15 s

94°C/5min 55°C/15 s 72°C/10 s

(×1) 72°C/ 5min

Size of in product basepairs 273

A = Adenine C = Cytosine G = Guanine T = Thymidine

Fig. 1. Restriction fragment length polymorphisms to detect the E6 nt 130 A to G variant.

18


8. Aerosol-resistant tips. 9. DNase-free plastics (Elkay Laboratory Plastics, Ltd., UK).

2.3. RFLP 1. 2. 3. 4.

Phenol:chloroform:isoamylalcohol (50%:48%:2% v/v). Absolute ethanol, and 70% v/v aqueous ethanol. Restriction endonucleases and buffers: Xcm1, Ssp1, Nsp1, Nsi1, and Msp1. Bovine serum albumin (Sigma).

2.4. Agarose Gel Electrophoresis 1. 10X Orange G loading buffer: 30% (w/v) Ficoll (Sigma), 250 mM ethylenediamine tetraacetic acid (EDTA; disodium salt), 0.25% (w/v) Orange G (BDH, Ltd.). 2. Agarose, UltraPure, electrophoresis grade (Invitrogen). 3. Tris-borate EDTA (TBE) buffer: 0.9 M Tris, 0.9 M boric acid, 2 mM EDTA. 4. Ethidium bromide: 10 mg/mL stock solution. 5. Molecular-weight size marker (e.g., 1 kb Ladder).

2.5. DNA Sequencing 1. 2. 3. 4.

Qiagen™ columns for purifying amplicons. pGem (Promega). Escherichia coli JM109 cells (Promega). Commercially obtainable T7 and SP6 primers (Sequenase™ kit, Pharmacia, Ltd.).

3. Methods 3.1. Preparation of Clinical Samples for PCR 1. Use PBS to resuspend cells from cell-lines or cervical brush smears in four 1-mL aliquots for PCR. 2. Centrifuge a 1-mL sample at 10,000g, resuspend the cell pellet in 200 µL PK solution, and incubate overnight at 55°C. 3. Inactivate the PK (by heating the sample to 90°C for 10 min), then store at –70°C.

3.2. Polymerase Chain Reactions 3.2.1. General PCR Considerations 1. Stringent precautions must be taken to prevent false-positives as a result of contamination, as described in Chapter 23, Subheading 3.1.

3.2.2. E5 PCR 1. Use a 20-µL aliquot of the PK-treated cell suspension directly in a 200-µL volume PCR using a proofreading DNA polymerase to amplify nt 3866 to 4077 of the HPV-16 genome (see Note 1, refs. 14,15, and Fig. 2). 2. For each batch of 20 clinical samples, two negative controls (molecular biology– grade water and A431 cells) and two positive controls (either pAt-16 plus CaSki


19

Fig. 2. Restriction polymorphism fragment analysis of human papillomavirus (HPV) type 16 E5. or pAt-16 plus SiHa DNA) should be included. The PCR is performed using the conditions shown as a guideline (Table 1); if problems are encountered, re-optimize conditions (see Note 2). 3. After PCR amplification, add 100 µL of MBG dH2O and then extract the DNA using phenol:chloroform. Precipitate the DNA with 1 mL of absolute ethanol, wash with 70% v/v aqueous ethanol, and then dry under vacuum for 10 min. Finally, resupsend the pellets in 20 µL of sterile MBG dH2O and store at –20°C prior to restriction digests.

3.2.3. E5 RFLP The E5 PCR amplifies HPV-16 E5 wild-type gene between nucleotides (nt) 3866 to 4077 and has been used previously to identify particular variants most commonly associated with cervical disease (16). For the reference HPV-16 sequence, this region contains over 45 restriction endonuclease (RE) cleavage sites, at least three of which (Xcm I [3872CCANN NNN↓NNNNTGG3886: where N = any nt]; Ssp I [nt 3978AAT↓ATT3983]; and, Nsp I [4077ACATG↓C4082]) are disrupted in different reported HPV-16 variants (6) (Fig. 2). These RE can therefore be used in an RFLP assay to identify eight HPV-16 variants (see Table 2 and Note 3).

20

Cason, Bible, and Mant Table 2 Designation of E5 Restriction Fragment Length Polymorphism Patterns Variant 1 2 3 4 5 6 7 8

XcmI

SspI

NspI

+ + – + – + – –

+ – + + – – + –

+ + + – + – – –

1. Resuspend DNA pellets in 50 µL of MBG water. 2. Prepare four 5-µL aliquots for each sample, three of which are subjected to an overnight digestion at 37°C with 5 U of one RE (SspI, XcmI, or NspI). 3. Separate the products by agarose gel electrophoresis (Subheading 3.2.4.). 4. Spiking experiments indicate that mixed infections can be identified in the RFLP assay only when the least frequent variant exceeds 10% (w/w) of the total HPV16 E5 DNA present (Fig. 3). In our experience, all eight possible RFLP patterns have been detected among samples from patients, with RFLP patterns 1 (28%), 2 (32%), and 9 (36%) being most common among cytologically normal women infected with HPV 16 (16). In HPV-16-positive women with cervical neoplasia, pattern 2 was present in 63% of cases and pattern 1 in 21%.

3.2.4. Agarose Gel Electrophoresis of PCR Products/RFLP Digests 1. Prepare a 2% (w/v) agarose gel using TBE buffer and 5 µL ethidium bromide solution per 100 mL agarose gel. 2. Mix PCR amplicons or RFLP digests with 10X Orange G loading buffer and electrophorese at 125 V for approx 1 h. Ensure molecular-weight standards are run in parallel with the molecular-weight markers analyzed on the gel to assess the size of PCR products. 3. Visualize bands on a ultraviolet transilluminator and compare amplicon size against molecular-weight marker. Photographic records should be obtained.

3.3. RFLP Detection of A→G Variation at Position 131 Within the E6 Open Reading Frame In this section we describe two RFLP assays to detect the E6 variant (A→G at position 131) described by Ellis et al. (17), which may be highly associated with women with high-grade CIN and with human leukocyte antigen B7 gene.


21

Fig. 3. Sensitivity of the human papillomavirus (HPV) type 16 restriction fragment length polymorphism to mixtures of variants. For this spiking experiment, each group of four lanes corresponds to (left to right) undigested amplicon, SspI, digest, XcmI digest, and NspI digest. A1: prototypic HPV16 pAt-16. A2-C10: mixtures of prototypic and HPV16 variants in different ratios from 90% prototypic/10% variant (A2) through to 10% prototypic/90% variant (C10). C11: 100% variant. M: molecular weight marker (sizes in bp).

3.3.1. E6 PCRs The first E6 PCR uses the primers E61A and E62A (18); this amplifies a 469-bp fragment of the E6 gene (Fig. 1A). The second primer set consists of E61B and E62B, which amplifies a 234-bp fragment of E6 (Fig. 1B).

3.3.2. E6 RFLP 1. Following PCR amplification with primer set A, digest 5 µL of product with 10 U NsiI (see Note 3) in a final dilution of 20 µL of buffer at 37°C overnight. 2. Further digest 10 µL of this reaction with Msp-I under the same conditions except for the addition of bovine serum albumin at 0.2 g/L. 3. PCR products produced using primer set B are digested with MspI alone. 4. Separate products on a 1.5% agarose gel containing ethidium bromide, and then visualize by trans-illumination with ultraviolet light.

22


5. Variants can be identified by the presence of the unique MspI site. Thus, in the case of the PCR using primer set A, the 469-bp product is digested with NsiI to produce fragments of 163 and 306 bp. These products are then digested with MspI, which cuts the 306-bp fragment into 245- and 61-bp products and cuts the 163-bp fragment into 127 and 36 bp only when the variant is present. The variant can be seen after electrophoresis, where the 163-bp fragment of the prototypic sequence is replaced by a 127-bp fragment, which indicates the variant is present. In the case of amplicons produced using primer set B, MspI digestion results in a band of 234 bp when the prototypic sequence is present, or a band of 192 bp when the nt 131 E6 variant is present.

3.3.3. DNA Sequencing There are two main approaches for DNA sequencing analysis. First, one can perform direct/bulk sequencing of PCR amplicons, which will provide a consensus of any HPV-16 sequences present; second, a more accurate approach is to clone the PCR products into a plasmid and then sequence >20 clones. 1. Purify E5 PCR amplicons using Qiagen columns and blunt end clone into pGem. 2. Add transformed plasmids into E. coli JM109 cells and grow overnight at 37°C on agar plates. 3. Select white colonies, grow midi-cultures, and purify plasmid DNA using Qiagen columns. 4. Sequence the inserts in both orientations (with T7 and SP6 primers that recognize sequences in pGem which flank the inserted E5 DNA) using Sequenase kit. 5. Analyze resulting E5 products representing HPV-16 E5 DNA sequence between nucleotides 3866 and 4077 on a DNA sequencer.

3.3.4. DNA Sequence Analyses A variety of free-to-use software is available on the net for DNA sequence analyses (Table 3). DNA sequence data can be easily arranged into a FASTA format: >DNA sequence title (hard return) ACCGGGGGGGGTGCTCAG . . . (but containing no “hard returns”)

This file can be saved as a normal text “.doc” file and manipulated for sequence editing (e.g., use the “Find” option in the “Edit” function of Microsoft Word to rapidly identify the primer sites, and the “Replace” function to remove hard returns). Such FASTA files can then be cut and pasted into an alignment program such as CLUSTAL-W (http://www.ebi.ac.uk/ clustalw/index.html or, http://clustalw.genome.ad.jp/). These programs will also permit “gap stripping” so that only like sequences of DNA are compared, the production of homology measurements, bootstrap analyses (which indicate how many times a given


23

Table 3 Useful Web Sites for DNA Analyses http://cti.itc.virginia.edu/~cmg/Demo/wheel/wheelApp.html http://www.ncbi.nlm.nih.gov/BLAST/ http://searchlauncher.bcm.tmc.edu/multi-align/multi-align.html http://ca.expasy.org/ http://www.ddbj.nig.ac.jp/E-mail/clustalw-e.html http://www.ualberta.ca/%7Estothard/javascript/rev_comp.html http://pbil.univ-lyon1.fr/alignment.html http://www.cbs.dtu.dk/databases/PhosphoBase/predict/predict.html http://www.ch.embnet.org/software/TCoffee.html http://www.ebi.ac.uk/clustalw/index.html http://www.ebi.ac.uk/services/ http://clustalw.genome.ad.jp/

branch occurs when trees are produced with a random input of the sequence files), and the drawing of phylogenetic trees. The trees can then be pasted into TreeView (http://taxonomy.zoology.gla.ac.uk/rod/treeview.html) in order to obtain phylogenetic trees that will indicate the amount of variation. 4. Notes 1. Phenol/chloroform extraction of target DNA often results in significant loss of target material. For this reason, we habitually do not use this method and just treat samples with proteinase K. To check that no significant carry-over of potential PCR inhibitors has occurred, all samples should also be tested by PCR for a “housekeeping” gene such as β-globin. 2. For all PCRs, it is recommended that “checkerboard” analyses (titrations of differing concentrations of different PCR components) be performed to optimize the PCRs by checking the optimal concentrations of primers, dNTPs, and magnesium ions. 3. There can be considerable variation in the efficacy of purportedly the same enzymes obtained from different suppliers; however, in our experience NEB, Ltd., produces highly active RE. Prior to any RFLP analyses of clinical samples, the activities of any RE should be determined by digesting a known section of DNA containing the appropriate RE cut sites.

References 1. zur Hausen, H. (2002) Papillomaviruses and cancer: from basic studies to clinical application. Nat. Rev. Cancer 2, 342–350. 2. zur Hausen, H. (2001) Oncogenic DNA viruses. Oncogene 20, 7820–7823. 3. Cavuslu, S., Goodlad, J., Connor, A., et al. (1997) Relationship between human papillomavirus infection and overexpression of p53 protein in cervical carcinomas and lymph node metastases. J. Med. Virol. 53, 111–117.

24


4. IARC (1995) Epidemiology of infection. In, IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Human Papillomaviruses, Lyon: International Agency for Research on Cancer, World Health Organisation, pp. 60–65. 5. Conrad-Stoppler, M., Ching, K., Stoppler, H., Clancy, K., Schlegel, R., and Icenogle, J. (1996) Natural variants of the human papillomavirus type 16 E6 protein differ in their abilities to alter keratinocyte differentiation and to induce p53 degradation. J. Virol. 70, 6987–6993. 6. Chan S-Y., Ho, L., Ong, C-K., et al. (1992) Molecular variants of human papillomavirus type 16 from four continents suggest ancient pandemic spread of the virus and its coevolution with mankind. J. Virol. 66, 2057–2066. 7. Fujinaga, Y., Okazawa, K., Nishikawa, A., et al. (1994) Sequence variation of human papillomavirus type 16 E7 in preinvasive and invasive cervical neoplasias. Virus Genes 9, 85–92. 8. Hecht, J. L., Kadish, A. S., Jiang, G., and Burk, R. D. (1995) Genetic characterization of the human papillomavirus (HPV) 18 E2 in clinical specimens suggests the presence of a subtype with decreased oncogenic potential. Int. J. Cancer 60, 369–376. 9. Londesborough, P., Ho, L., Terry, G., Cuzick, J., Wheeler, C., and Singer, A. (1996) Human papillomavirus genotype as a predictor of persistence and development of high-grade lesions in women with minor cervical abnormalities. Int. J. Cancer 69, 364–368. 10. Xi, L-F., Koutsky, L. A., Galloway, D. A., et al. (1997) Genomic variation of human papillomavirus type 16 and risk for high grade cervical intraepithelial neoplasia. J. Natl. Cancer Inst. 89, 796–802. 11. Zehbe, I., Wilander, E., Delius, H., and Tommasino, M. (1998) Human papillomavirus 16 E6 variants are more prevalent in cervical carcinoma than the prototype. Cancer Res. 58, 829–833. 12. Seedorf, K., Kraemmer, G., Duerst, M., Suhai, S., and Rowekamp, W. G. (1985) Human papillomavirus type 16 DNA sequence. Virology 145, 181–185. 13. Halbert, C. L. and Galloway, D. A. (1998) Identification of the E5 open reading frame of human papillomavirus type 16. J. Virol. 62, 1071–1075. 14. Cavuslu, S., Starkey, W. G., Kaye, J. N., et al. (1996) Detection of human papillomavirus type-16 (HPV-16) DNA utilising microtitre-plate based amplification reactions and a solid-phase enzyme-immunoassay detection system. J. Virol. Methods 58, 59–69. 15. Mant, C., Kell, B., Best, J. M., and Cason, J. (1997) Polymerase chain reaction protocols for the detection of DNA from mucosal human papillomavirus types 6,-11, -16, -18, -31 & -33. J. Virol. Methods 66, 169–178. 16. Bible, J. M., Mant, C., Best, J. M., et al. (2000) Cervical lesions are associated with human papillomavirus type 16 intratypic variants that have high transcriptional activity and increased usage of common mammalian codons. J Gen. Virol. 81, 1517–1527.


25

17. Ellis, J. R. M., Keating, P. J., Baird, J., et al. (1995) The association of an HPV 16 oncogene variant with HLA B7 has implications for vaccine design in cervical cancer. Nat. Med. 1, 464–470. 18. Luxton, J., Mant, C., Greenwood, B., et al. (2000) HPV16 E6 oncogene variants in women with cervical intraepithelial neoplasia. J. Med. Virol. 60, 337–341.

Analysis of Cutaneous HPV-Associated Lesions

27

3 Histochemical Analysis of Cutaneous HPV-Associated Lesions Kiyofumi Egawa Summary Hematoxylin and eosin (H&E) staining of cutaneous warts is presented to illustrate the practical methods utilized for histochemical analysis of cutaneous human papillomavirus-associated lesions. Every step of the staining procedure, from sampling of the specimens to microscopic examination of the stained sections, is detailed with reference to the recent achievements in this field.

1. Introduction Histochemistry is a biological approach that permits a precise interpretation of the chemistry of cells and tissues in relation to structural organization. Hematoxylin and eosin (H&E) stain is the most widely used method in histochemical analysis (1–4) of cutaneous human papillomavirus (HPV)-associated lesions. Several important aspects of HPV cell biology and virology have been disclosed utilizing this method either alone or in combination with the other methods of immunohistochemistry, cell kinetics, or molecular cell biology. Two important aspects of the nature of this group of heterogeneous viruses are the way in which specific HPV genotypes are associated with distinct clinical and histological morphologies and the way specific HPV genotypes affect distinct anatomical sites. The former is best evidenced by the HPV-type specific cytopathic or cytopathogenic effect (CPE) (5–13), whereas the latter is suggested by the marked preference of each HPV genotype for specific tissues and sites (5–8). Recent studies have also suggested that specific HPV genotypes may target epithelial stem cells at specific anatomical sites (14–17). In this chapter, I illustrate the practical methods of H&E staining. These include the sampling of specimens, and their fixation, embedding, sectioning, From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

27

28

Egawa

staining, and microscopic examination. I have included references to recent achievements in HPV-associated cutaneous pathology to illustrate the methods. 2. Materials 1. 2. 3. 4.

5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

16. 17. 18. 19. 20. 21.

Local anesthetics. Scalpel. Plastic containers of various sizes (Fig. 1A). Fixative solution: 10% neutral buffered formalin (pH 7.0), concentrated (40%) formaldehyde solution (100 mL), distilled water (900 mL), acid sodium phosphate (monohydrate) (4 g), anhydrous disodium phosphate (6.5 g). Xylene (100%). Ethanol (70%, 80%, 90%, 95%, 100%). Paraffin wax—e.g., Histosec (Merck Co.). Cassettes or molds of various sizes—e.g., Tissue-Tek Processing/Embedding Cassette (Sakura Finetek Co.) (Fig. 1A,B). Microtome. Glass microscope slides. Water bath. Hot plate. Staining basket—e.g., Matsunami Glass Ind. (Fig. 2). Staining vat—e.g., Matsunami Glass Ind. (Fig. 2). Carazzi hematoxylin (modified): hematoxylin crystals (1.5 g) in 10% alcohol (10 mL), aluminium potassium sulfate 12-hydrate (50 g) in distilled water (800 mL), sodium iodide (0.3 g), glycerol (200 mL), acetic acid (15 mL). 0.25% aqueous solution of hydrochloric acid. Alcoholic eosin solution: water-soluble eosin Y (5.0 g), distilled water (500 mL), 80% ethanol (1500 mL). Mounting medium—e.g., Malinol (Muto Pure Chemical Co.). Glass coverslips. Light microscope. Automatic processor—e.g., Vacuum Infiltration Processor (Sakura Finetek Co.) (optional).

3. Methods The methods described below outline (1) sampling of the specimens, (2) fixation, (3) embedding and sectioning, (4) hematoxylin and eosin stain, and (5) histopathological examination. Fig. 1. (opposite page) (A) Plastic containers (Co), cassettes (C) and molds (M), of different sizes. (B) A tissue specimen placed into a mold (M) holding the melted paraffin. Inset: A paraffin block mounted on a plastic cassette. (C) A cut paraffin section is floated on water and scooped up using a glass microscope slide. The section is allowed to stretch by placing it in warm water and is then mounted up again by being scooped up onto a glass microscope slide.


29

30

Egawa

Fig. 2. Deparaffinization, rehydration, and staining. B: staining basket; V: staining vat containing xylene or graded ethanol; S: glass microscope slide; M: mounting medium in a glass bottle; C: glass cover slip.

3.1. Sampling of the Specimens In most instances, an incisional or excisional biopsy is taken from a welldeveloped typical lesion by scalpel or punch under local anesthesia. However, in research, consideration should be given to the selection of the lesions to obtain the specimens which are most appropriate for the aim of each study. In the presented case, a biopsy is taken from a very early lesion of a plantar wart to evaluate the histological localization of the initial changes of HPV infection in terms of its association with epidermal stem cells (16–19) (see Note 1). Tissues submitted for histology must not be more than 5 mm thick and not larger than the dimensions of the cassette used; otherwise, they will not be adequately fixed or infiltrated by paraffin.

3.2. Fixation 1. Following removal, place the biopsy specimen immediately into a container containing at least 10 volumes of fixative solution, ensuring that it completely surrounds the specimen on all sides (Fig. 1A). 2. Allow adequate time for fixation: the minimum period for a specimen 4 mm thick is 8 h (4). We usually carry out the fixation overnight at room temperature (see Note 2).


31

3.3. Embedding and Sectioning 3.3.1. Dehydration, Clearing, and Infiltration 1. Remove the water from the tissue (dehydration) by immersing the formalin-fixed specimens into graded ethanol from 70% to absolute at room temperature: 70% (12 h), 80% (12 h), 90% (12 h), 95% (12 h), and 100% (12 h, three times). This is necessary because the water contained in the tissue is not miscible with paraffin. 2. Pass the tissue through two or three changes of xylene until all the alcohol is replaced by xylene (clearing): 100% (60 min, three times). 3. Place the tissue in melted paraffin at 60°C until all the xylene has been replaced by paraffin (infiltration).

These steps can be carried out by an automatic processor.

3.3.2. Embedding The current dermatopathology has been established on the morphologic descriptions of skin sections cut perpendicular to the skin surface (1–4) (Fig. 3 faces A,B). Place the tissue into a mold (Fig. 1A), holding the melted paraffin (Fig. 1B) in such a manner that the tissue will be cut perpendicular to the skin surface when it is mounted on a microtome (embedding) (see Note 3). In the case presented here, further attention is paid to the precise orientation for embedding; to allow examination of the initial histological changes associated with HPV infection in the shallow or deep epidermal ridges (16–19) (see Note 1 and Fig. 3, face A), the specimen is placed so that the cut will be perpendicular to the epidermal ridges. When the specimen is cut in parallel with ridges (Fig. 3, face B), it is impossible to evaluate such association (see Note 3).

3.3.3. Sectioning 1. After cooling for several minutes to an hour at room temperature (depending on the size), the hardened paraffin blocks are mounted onto a cassette (Fig. 1B) and placed into the microtome in such a manner that the edge of the knife does not hit the horny layer first (see Note 4). 2. Cut the tissue into 4-µm thick slices using a microtome (sectioning). 3. Float the cut paraffin sections on water at room temperature. Scoop it up onto a glass microscope slide, ensuring the section is smooth against the slide (Fig. 1C). Move the slide into warm water (approx 48°C); the section will float off; leave for a few seconds to stretch out and remove any creases. To mount up the section, scoop it up again onto the glass slide (Fig. 1C), and dry on a hotplate (approx 42°C) for 10 min or more to allow the section to adhere to the glass slide.

3.4. Hematoxylin and Eosin Stain In H&E staining, a specific chemical-identifying reaction is achieved utilizing hematoxylin as a basic dye and eosin as an acidic dye, to identify the chemi-

32

Egawa

Fig. 3. Nonlabeled: Drawing of the plantar skin tissue showing different faces to be cut corresponding to the aim of each examination: “A” represents the face cut perpendicular not only to the skin surface but also to the surface ridges and sulci; “B” represents the face cut perpendicular to the skin surface but in parallel with surface ridges; and “C” represents the face cut in parallel with the skin surface (horizontal section). (A–C): H&E-stained sections obtained from a minimal wart. A corresponds to Face A; B corresponds to Face B; and C corresponds to Face C of the drawing. Sections obtained from a central portion of the minimal wart, clearly demonstrate that the initial human papilloma virus-associated histological changes are restricted to the deep epidermal ridge in association with an eccrine duct (arrow) in A and C, whereas it is impossible to evaluate the initial histological changes in terms of its association with such epidermal architecture in B. R, surface ridges; S, sulci; W, a minimal wart; DR, deep ridges; SR, shallow ridges; ED, eccrine ducts; arrows, eccrine ducts.

cal substances contained in the cells or tissues. Hence, a cellular or tissue component that binds hematoxylin is described as being basophilc; conversely, a component that binds eosin is acidophilic or eosinophilic. After the nucleus is stained blue to purple with hematoxylin, as a result of its nucleic acid content, the cytoplasm is counterstained pink to red with eosin, mainly as a result of its structural proteins (1–4).


33

3.4.1. Deparaffinization and Rehydration Before the sections can be stained, the paraffin permeating them has to be removed and replaced by water (Fig. 2). 1. Dip the sections adhering to glass slides first into xylene (10 min, two changes) to remove the paraffin and then into absolute ethanol to remove the xylene (5 min, two changes). 2. Then pass the sections through ethanol of decreasing strength and finally distilled water: 90% ethanol (five dips), 80% ethanol (five dips), 70% ethanol (five dips), and distilled water (a few minutes).

3.4.2. Staining Procedure 1. Stain with Carazzi hematoxylin (15 min); rinse in tap water (5 to 10 min). 2. Differentiate in 0.25% aqueous solution of hydrochloric acid (two to three dips); wash in running tap water (5 min). 3. Stain with eosin (15 s to 2 min, depending on the age of the eosin and the depth of counterstain desired). 4. Dehydrate: 70% alcohol (one dip), 80% alcohol (one dip), 90% alcohol (five dips), and absolute alcohol (five dips, three changes). 5. Clear in xylene (five dips, two changes, then until mounting), mount in medium, and cover with a cover slip.

3.5. Histopathological Examination The final and most important step of this method is the histopathological examination by light microscopy. In the examination, it is important to be familiar with not only the morphological and cell biological aspects of the normal human skin (3,4,19) but also HPV type-specific CPE (5–13) (Subheading 3.5.1.), on which every HPV-associated histopathological change should be analyzed.

3.5.1. HPV Type-Specific CPE HPV type-specific CPE is the central schema when we analyze and understand the cutaneous HPV-associated histopathology. The concept was first suggested by the characterization of distinct HPVs from different types of cutaneous warts (5–8,20,21): HPV 1 from deep plantar warts (myrmecia or inclusion warts), HPV 2 from common warts, and HPV 3 from flat warts. A characteristic histological feature of HPV 1 infection is intracytoplasmic inclusion bodies (ICB) appearing as eosinophilic granules (granular type of inclusion body: Gr-ICB) in most of the cells of the epidermis (Fig. 4A). The cells infected with HPV 2 are described as often vacuolated, containing condensed heterogeneous keratohyaline granules (Fig. 4B). Highly characteristic of HPV 3-induced warts is perinuclear clarification around small basophilic,

34

Egawa

Fig. 4. Cytopathic effects of human papilloma virus (HPV) 1 (A), HPV 2 (B), and HPV 3 (C). Arrows: granular inclusion bodies (A), vacuolated cells (B), and bird’seye cells (C); arrow + K: condensed keratohyaline granules (B).

sometimes pyknotic (pyknosis: shrinking of nuclei), usually centrally located nuclei and well-defined borders of the cells (so-called bird’s eye cells) (Fig. 4C). The CPE results from the derangement of terminal differentiation of keratinocytes infected by HPV (5,6). The importance of characterization of CPE is best evidenced in a group of novel inclusion warts (5–13).

3.5.2. Inclusion Warts For some time, the term inclusion wart has been used synonymously with myrmecia or HPV 1-induced warts (5–10,20,21). This is because the myrmecia was the first established clinical entity in which specific Gr-ICBs were identified (22) (Fig. 4A), and further characterized as HPV 1-induced inclusion bodies (5–10,20,21). However, careful clinicopathological examinations reveal that two novel types of inclusion warts exist (9,10)—punctate warts containing a heavily stained intracytoplasmic tonofibril-like substance within which there are filamentous structures (filamentous type of ICB: Fl-ICB) (9,10) (Fig. 5A) and pigmented warts containing an eosinophilic, homogeneous substance in each cell (homogeneous type of ICB: Hg-ICB) (9,10,13) (Fig. 5B,C,D). Similar Hg-ICBs are also identified in two other new types of skin lesions—i.e., the cystic papilloma (11) and the ridged wart (12). Cystic papilloma (or wart) is a new clinical entity, which includes HPV-associated epidermoid cysts (6,8,11) (see Note 5). These unusual CPEs, which have not previously been described, lead to the conclusion that the warts could be induced by novel types of HPV (9–11). Indeed three new types, HPV 60, 63, and 65, have been cloned from such


35

Fig. 5. Filamentous type inclusion bodies (Fl-ICB) associated with human papillomavirus (HPV) 63 (A) and homogeneous type ICBs (Hg-ICBs) associated with HPV 4 (B), HPV 60 (C), and HPV 65 (D). Arrows: intracytoplasmic inclusion bodies.

lesions (9,23). The direct associations are currently observed among myrmecia, Gr-ICB, and HPV 1; among punctate warts, Fl-ICB and HPV 63; and among pigmented warts, Hg-ICB and the related types of HPV—HPV 4, 60, and 65 (6–10,13) (see Note 6). HPV 60 was originally identified in (11) and cloned (23) from HPV-associated epidermoid cysts with the Hg-ICB. One does not always get what one expects when staining (and often not what is already documented in the literature). For example, a punctate wart

36

Egawa

with Fl-ICBs may be expected to contain HPV 63 DNAs, considering the HPV-type specific CPE. However, in such a lesion, it may be found following molecular analysis that each single cell contains HPV 1 as well as HPV 63 (24) (see Note 7). In the case presented here, an eccrine-centered distribution of the initial CPE in the deep epidermal rete ridge was suggested from serial sections obtained from the entire biopsy specimen (16,17) (see Fig. 3). It has also been suggested from recent histopathological analyses using serial sections that the HPV-associated epidermoid cysts may develop from an eccrine duct by HPV infection (25,26). To produce reliable histopathological information, we recommend that histochemical analysis be utilized in combination with other methods such as immunohistochemistry, cell kinetics (27), or molecular cell biology, and that as many serial sections as possible be taken from each specimen (1,16,17,25,26,28). 4. Notes 1. Evidence has suggested that the epidermal stem cells exist in the bulge region of the hair follicle in hair-bearing skin, while they exist in the basal layer at the deepest point of the deep rete ridges in non-hair-bearing palmoplantar skin (16–19). Schmitt et al. (14) and Boxman et al. (15) propose the idea that HPVs may primarily target the epidermal stem cells in hair follicles in hair-bearing skin. In non-hairy palmoplantar skin, alternating ridges and sulci are present on the surface, with ridges and sulci corresponding histologically to the deep and shallow epidermal ridges, respectively. To address the question as to whether HPVs also target epidermal stem cells in the palmoplantar skin, a very early lesion is required whose size is small enough to estimate the correlation between the wart and ridges (16,17) (Fig. 3). 2. Tissues should not be frozen once they have been placed in the fixative solution, or a peculiar ice crystal distortion will result. In textbooks of histopathology, it is recommended that, during winter or in countries with cold weather, either 95% ethanol, 10% by volume, be added to the formalin solution or the specimen be allowed to stand in the formalin solution at room temperature for at least 6 h to prevent such artifacts (1–4). However, one should be aware that some CPEs might be lost from cells in fixatives that contain alcohol, resulting in figures of vacuolated cell-like, koilocytic, or signet ring cell-like appearance (Fig. 6). 3. This is not always the case in research. Although proper identification and orientation of the specimen is always important for the adequate histopathological evaluation of the lesion, specimens may also be cut with other deviations corresponding to the aim of each particular study. In addition, sometimes the histological sections deviate from the ordinary face A to B or C, even though this was not what was intended. 4. This opinion is the opposite of a current recommendation that specimens should be placed in such a manner that the edge of the knife hits the epidermis first. However, in wart specimens, it is actually quite common that an extremely


37

Fig. 6. Hematoxylin and eosin (H&E)-stained sections obtained from human papillomavirus (HPV) 1-induced myrmecia show marked cell vacuolization-like artifacts instead of granular type inclusion bodies (Gr-ICB), when specimens are fixed in 80% ethanol. hypertrophied hard horny layer disturbs our smooth sectioning when the specimens are placed in such a manner. 5. We propose to call the HPV-associated epidermoid cysts “cystic papilloma” in humans (11). This is because we follow the suggestion of Rous and Beard, who gave the name cystic papilloma to the epidermoid cysts produced in rabbits by the Shope papillomavirus (29). We assume that the cystic papilloma of rabbits corresponds to the HPV-associated epidermoid cyst in humans (11). 6. The Gr-ICB is mainly composed of HPV 1 E4 proteins (30). The Hg-ICB is also associated with HPV 4 or HPV 65 E4 proteins (5,13). Although the exact role of the E4 proteins has not yet been determined, interference with normal intermedi-

38

Egawa

ate filament assembly has been proposed as a function (30). Indeed, HPV 16 E4 protein is involved in causing the collapse of the keratin cytoskeleton (31). The heterogeneity of ICBs, and their association with specific (or the related) types of HPV, will provide a useful model system for studying the interaction between keratinocytes and HPVs, especially in the functional aspect of E4 gene expression (9,10,13,24). In this aspect, it is interesting that the related types of HPV, HPV 4, HPV 60, and HPV 65 induce a very distinct CPE—i.e., Hg-ICB. 7. The double infection with HPV 1 and HPV 63 within a single cell is of special interest, because of its HPV 1-type CPE (6,8,24). This poses the very important problem of a possible interference between the viruses or a role of one virus in transactivation of the other viruses (6,8,24). We would emphasize that CPE is not a mere diagnostic marker for recognizing HPV type, but an important natural representation of the tight association between the genotype and phenotype. Any deviations from the basic association should suggest the underlying virus- or hostrelated factors that could influence CPE.

Acknowledgments I acknowledge Drs. Yumi Honda, Ethel-Michele de Villiers, Hidero Kitasato, John Doorbar, Tomas Iftner, and Harald zur Hausen for the excellent collaborations and fruitful discussions; Mrs. Chiemi Shiotsu for technical assistance; Dr. Clare Davy for helping in preparation of the manuscript; and Mrs. Motoe Egawa for encouragement. References 1. Ham, A. W. and Cormack, D. H. (1979) Histology, its place in the biological and medical sciences, and how it is studied. In: Histology, 8th ed, Ham, A. W. and Cormack, D. H. (eds.), J. B. Lippincott Company, Philadelphia, pp. 3–32. 2. Rosai, J. (1996) Gross techniques in surgical pathology. In: Ackerman’s Surgical Pathology 8th ed., vol. 1, Rosai, J. (ed.), Mosby–Year Book, Inc., St. Louis, MO, pp. 13–28. 3. Mehregan, A., Hashimoto, K., Mehregan, D., and Mehregan, D. (1995) Technical data, including pitfalls and artifacts. In: Pinkus’ Guide to Dermatopathology, 6th Ed., Mehregan, A., Hashimoto, K., Mehregan, D., Mehregan, D. (eds.), Appleton and Lange, East Norwalk, CT, pp. 49–68. 4. Lever, W. F. and Schaumburg-Lever, G. (1990) Laboratory methods. In: Histopathology of the Skin, 7th ed, Lever WF, Schaumburg-Lever G (eds.), J. B. Lippincott Company, Philaderphia, pp. 44–54. 5. Croissant, O., Breitburd, F., and Orth, G. (1985) Specificity of cytopathic effect of cutaneous human papillomaviruses. Clin. Dermatol. 3, 43–55. 6. Gross, G. E. and Jablonska, S. (1997) Skin warts: gross morphology and histology. In: Human Papillomavirus Infections in Dermatovenereology, Gross, G. E. and von Krogh, G. (eds.), CRC Press, Boca Raton, FL, pp. 243–258.


39

7. Gross, G. E., Jablonska, S., and Huegel, H. (1997) Skin: diagnosis. In: Human Papilloma Virus Infection: A Clinical Atlas, Gross, G. E. and Barrasso, R. (eds.), Ullstein Mosby, Berlin, pp. 63–123. 8. Syrjaenen, K. and Syrjaenen, S. (2000) HPV infections of the skin. In: Papillomavirus Infections in Human Pathology, Syrjaenen, K. J. and Syrjaenen, S. M. (eds), John Wiley & Sons, Ltd., Chichester, UK, pp. 315–340. 9. Egawa, K., Delius, H., Matsukura, T., Kawashima, M., and de Villiers, E-M. (1993) Two novel types of human papillomavirus, HPV 63 and HPV 65: comparison of their clinical and histological features and DNA sequences to other HPV types. Virology 194, 789–799. 10. Egawa, K. (1994) New types of human papillomaviruses and intracytoplasmic inclusion bodies: a classification of inclusion warts according to clinical features, histology and associated HPV types. Br. J. Dermatol. 130, 158–166. 11. Egawa, K., Inaba, Y., Ono, T., and Arao, T. (1990) “Cystic papilloma” in humans?—demonstration of human papillomavirus in plantar epidermoid cysts. Arch. Dermatol. 126, 599–1603. 12. Honda, A., Iwasaki, T., Sata, T., Kawashima, M., Morishima, T., and Matsukura, T. (1994) Human papillomavirus type 60-associated plantar wart: ridged wart. Arch. Dermatol. 130, 1413–1417. 13. Egawa, K., Honda, Y., Inaba, Y., and Ono, T. (1998) Pigmented viral warts: a clinical and histopathological study including human papillomavirus typing. Br. J. Dermatol. 138, 381–389. 14. Schmitt, A., Rochat, A., Zeltner, R., et al. (1996) The primary target cells of the high-risk cottontail rabbit papillomavirus colocalize with hair follicle stem cells. J. Virol. 70, 1912–1922. 15. Boxman, I. L. A., Berkhout, R. J. M., Mulder, L. H. C., et al. (1997) Detection of human papillomavirus DNA in plucked hairs from renal transplant recipients and healthy volunteers. J. Invest. Dermatol. 108, 712–715. 16. Egawa, K. (2003) Do human papillomaviruses target epidermal stem cells? Dermatology 207, 251–254. 17. Egawa, K. (2005) Eccrine-centered distribution of HPV 63 infection in the epidermis of the plantar skin. Br. J. Dermatol. in press. 18. Lavker, R. M. and Sun, T.-T. (2000) Epidermal stem cells: properties, markers, and location. Proc. Natl. Acad. Sci. USA 97, 13,473–13,475. 19. Holbrook, K. A. and Wolff, K. (2001) The structure and development of skin. In: Dermatology in General Medicine, 5th ed., Fitzpatrick, T. B., Eisen, A. Z., Wolff, K., Freedberg, I. M., and Austen, K. F. (eds.), McGraw-Hill, Inc., New York, pp. 70–141. 20. Gross, G., Pfister, H., Hagedorn, M., and Gismann, L. (1982) Correlation between human papillomavirus (HPV) type and histology of warts. J Invest. Dermatol. 78, 160–164. 21. Jablonska, S., Orth, G., Obalek, S., and Croissant, O. (1985) Cutaneous warts. Clinical, histologic, and virologic correlations. Clin. Dermatol. 3, 71–82.

40

Egawa

22. Lyell, A. and Miles, J. A. R. (1951) The myrmecia. A study of inclusion bodies in warts. Br. Med. J. 28, 912–915. 23. Matsukura, T., Iwasaki, T., and Kawashima, M. (1992) Molecular cloning of a novel human papillomavirus (type 60) from a plantar epidermoid cyst with characteristic pathological changes. Virology 190, 561–564. 24. Egawa, K., Shibasaki, Y., and de Villiers, E.-M. (1993) Double infection with human papillomavirus 1 and human papillomavirus 63 in a single cell of a lesion displaying only an human papillomavirus 63-induced cytopathogenic effect. Lab. Invest. 69, 583–588. 25. Egawa, K., Honda, Y., Inaba, Y., Ono, T., and de Villiers, E.-M. (1995) Detection of human papillomaviruses and eccrine ducts in palmoplantar epidermoid cysts. Br. J. Dermatol. 132, 533–542. 26. Egawa, K., Egawa, N., and Honda, Y. (2005) Human papillomavirus-associated plantar epidermoid cyst resulting from an epidermoid metaplasia of eccrine duct epithelium: a combined histological, immunohistochemical, DNA-DNA in situ hybridization, and three-dimensional reconstruction analysis. Br. J. Dermatol. in press. 27. Egawa, K., Iftner, A., Doorbar, J., Honda, Y., and Iftner, T. (2000) Synthesis of viral DNA and late capsid protein L1 in parabasal spinous layers of naturally occurring benign warts infected with human papillomavirus type 1. Virology 268, 281–293. 28. Honda, Y., Egawa, K., Baba, Y., and Ono, T. (1996) Sweat duct milia–immunohistological analysis of structure and three-dimensional reconstruction. Arch. Dermatol. Res. 288, 133–139. 29. Rous, P. and Beard, J. W. (1935): The progression to carcinoma of virus-induced rabbit papillomas (Shope). J. Exp. Med. 62, 523–548. 30. Doorbar, J., Cambell, D., Grand, R. J. A., and Gallimore, P. H. (1986) Identification of the human papillomavirus-1a E4 gene products. EMBO J. 5, 355–362. 31. Doorbar, J., Ely, S., Sterling, J., Mclean, C., and Crawford, L. (1991) Specific interaction between HPV 16 E1-E4 and cytokeratins results in collapse of the epithelial cell intermediate filament network. Nature 352, 824–827.

Cervical Intraepithelial Neoplasia

41

4 Histological Analysis of Cervical Intraepithelial Neoplasia Michael Babawale, Rashmi Seth, Adam Christian, Wessam Al-Utayem, Ragini Narula, and David Jenkins Summary A wide interobserver variation is seen even among competent histopathologists in the routine diagnosis of cervical intraepithelial neoplasia (CIN). As a result, early detection of lowgrade CIN (CIN 1) lesions, in particular, remains a major challenge both in routine diagnosis and in cervical screening. In this chapter, the salient diagnostic features of human papillomavirus infection and CIN lesions are demonstrated.

1. Introduction Human papillomavirus (HPV) infection is now universally accepted as the most important risk factor for carcinoma of the cervix in women (1). There is poor reproducibility and variable diagnostic accuracy among histopathologists, largely because of failure to adhere to accepted diagnostic guidelines (2). These limitations have led to the search for more sensitive and specific biomarkers, such as p16INK4a, with the hope of improving the diagnostic accuracy of cervical intraepithelial neoplasia (CIN) lesions (3). However, routine hematoxylin and eosin (H&E) remains the gold standard of diagnosing HPV and CIN lesions in women. Hematoxylin is extracted from the tree Haematoxylin campechianum, now mainly cultivated in the West Indies. Hematoxylin itself is not a stain; it is its oxidation product, hematin, that is the natural dye. Hematin is usually combined with a mordant such as aluminum salts (cation) to improve its binding affinity to anionic tissue sites such as nuclear chromatin. Eosin is a xanthene dye with the ability to distinguish between the cytoplasm of different types of cells and connective tissue fibers. Eosin combines extremely well with hematoxylin and hence, it has become a universally acceptable stain in demonstrating the general histological architecture of tissue (4). From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

41

42

Babawale et al.

CIN is a widely accepted term describing a range of dysplastic changes occurring in the cervical epithelium. CIN1 corresponds to mild dysplasia, CIN2 to moderate dysplasia, and CIN3 to severe dysplasia or carcinoma in situ. The Bethesda classification, commonly used in the United States, groups cervical epithelial lesions into two categories. The first includes HPV lesions and CIN1, and is called low-grade squamous intraepithelial lesions (LGSIL). The second category, high-grade squamous intraepithelial lesions (HGSIL), includes CIN2 and CIN3. For high-risk HPV (HR-HPV), general consensus primers (GP5+/6+) are used to amplify the viral DNA from the biopsies. DNA is normally extracted and purified from tissue sections and amplified. The polymerase chain reaction (PCR) products can then be analyzed using enzyme-linked immunosorbent assay (ELISA) and a cocktail of HR-HPV probes as previously described (5). For HPV 16 genotyping, type-specific primers designed in the HPV 16 E7 region are used in conjunction with real-time PCR to obtain viral-load measurements. In this chapter, we demonstrate a method of using H&E staining in different grades of CIN compared to normal tissue, and then correlate the results with the presence of HR-HPV and HPV 16 E7 gene viral loads in the biopsy material using real-time quantitative PCR assay. 2. Materials 1. Harris’s hematoxylin, composed of 2.5 g hematoxylin, 25 mL absolute alcohol, 50 g potassium alum, and 1.25 g mercuric oxide in 500 mL distilled water. The hematoxylin is dissolved in the absolute alcohol and is then added to the alum that has previously been dissolved in the warm distilled water in a 2-L flask. The mixture is rapidly brought to the boil, and the mercuric oxide is then slowly and carefully added. The stain is rapidly cooled by plunging the flask into cold water or a container of ice blocks. 2. Eosin. 3. 1% acid alcohol (1 mL concentrated HCl + 700 mL ethanol + 300 mL deionized H2O). 4. Graded alcohol: 70%, 95%, and absolute (99.5%). 5. Xylene. 6. DPX (NUSTAIN, Nottingham, UK). 7. Glass slides. 8. Cover slips. 9. An optical microscope (Zeiss, W. Germany). 10. Real-time PCR instrument (Mx4000, Stratagene, UK). 11. GP5+ forward primer: 5' TTTGTTACTGTGGTAGATACTAC 3'. 12. GP5+ reverse primer: 5' GAAAAATAAACTGTAAATCATT 3'. 13. HPV 16E7 gene forward primer: 5' GATGAAATAGATGGTCCAGC 3'. 14. HPV 16E7 reverse primer: 5' GCTTCGGTTGTGCGTACAAAGC 3'.

Cervical Intraepithelial Neoplasia 15. 16. 17. 18. 19.

43

Betaglobin forward primer: 5' ACACAACTGTTCACTAGC 3'. Betaglobin reverse primer: 5' GAACCCAAGAGTCTTCTTCTCTCT 3'. DNA kit (DNA Easy Kit, Qiagen, UK). Standard HPV 16 DNA (Advanced Biotechnologies, Inc., UK). QuantiTect SYBR® Green PCR master mix kit (Qiagen).

3. Methods The methods described below outline (1) H&E staining techniques, (2) realtime quantitative PCR using consensus and HPV 16 type-specific primers, and (3) how to interpret the data obtained from the clinical biopsy specimen.

3.1. Hematoxylin and Eosin Staining (4) 1. Cut the tissue sections (5 µm) on glass slides, fix by keeping them on the hotplate for 15–20 min, and then dip in two consecutive changes of xylene to remove paraffin (see Note 1). 2. Rehydrate the tissue sections through downgraded alcohol to water prior to staining. First place in absolute (99.5%) ethanol (ethyl alcohol), second into 90% ethanol (and 10% water), and finally into 70% ethanol (and 30% water). 3. Dip the sections in Harris’s hematoxylin for 5–10 min. 4. Wash in running water for 5 min to remove the excess hematoxylin stain. 5. Differentiate in 1% acid alcohol for 5–10 s (see Note 2). 6. Wash again in running water for 2 min. 7. Counter-stain in 1% eosin for 5–10 min (see Note 3). 8. Wash in running water for 2 min to remove the excess eosin (see Note 4). 9. Dehydrate through upgraded alcohol and then xylene. Dip slides through ascending grades of ethanol, first in 70% ethanol (and 30% water), followed by 90% ethanol (and 10% water), then absolute (99.5%) ethanol, and finally xylene (see Note 5). 10. Clear in two changes of xylene. 11. Mount using DPX (see Note 6). 12. View and interpret the slides.

The above steps could be carried out manually or by automated staining machines.

3.2. HPV Detection and Typing Before the PCR can be set up, DNA extraction from the tissue sections needs to be performed using a Qiagen DNA extraction kit following the manufacturer’s instructions. 1. Suspend 30 µm sections after de-waxing (see Note 7) in 200 µL of ATL digestion buffer (from the DNA extraction kit). 2. Incubate the samples overnight at 56°C in the presence of proteinase K; vortex occasionally until the tissue is completely lysed. 3. Extract DNA using the mini-columns (all buffers and columns provided in the kit); pure DNA is eluted in AE buffer (provided in the kit).

44

Babawale et al.

Table 1 Summary of Representative 16/40 of Biopsies Used N

Number of cases

Histological grade

HR-HPV positivity

HPV -16E7 positivity

1

3

Normal

3/3

2/3

2

2

CIN1

2/2

1/2

3

4

CIN2

4/4

3/4

4

4

CIN3

4/4

2/4

5

3

Invasive Cancer

3/3

3/3

Viral load Very low (10 ng)

HR-HPV, high-risk human papillomavirus; CIN, cervical intraepithelial neoplasia.

4. Set up PCR for housekeeping gene (betaglobin) to test for the presence of amplifiable DNA. PCR conditions are standard, using SYBR green master mix, and primers are added at 10 pmol per 25 µL reaction. Start PCR using 40 cycles of 58°C for 1 min and then 72°C for 30 s. 5. Samples positive for betaglobin (generating a 205-bp fragment when run on agarose gel) are then subjected to HPV PCR using consensus primers to obtain overall HR-HPV positivity. HPV 16 type-specific PCR is carried out using HPV 16 E7 type-specific primers (see Chapter 6). 6. All of the samples positive for HR-HPV (giving a 108-bp product) are then subjected to HPV 16 type-specific PCR using real-time quantitative PCR technique (see Note 8) and SYBR Green dye (see Note 9). 7. Positive controls are set up with increasing dilutions of standard HPV 16 DNA to give a calibration graph. Negative controls are always set up by substituting deionized water for the template. The HPV 16 E7 assay is sensitive (0.0001 ng/ tube) and specific for HPV 16 only, and there is no cross-reaction with any other HPV types when tested. Samples that are positive will also have viral load measurements (in ng/tube). 8. Interpret the H&E staining with grade of CIN and viral load measurements. Our results have shown that there is a significant positive correlation between CIN grade and HPV viral load (see Table 1).

3.3. Interpretation of Results Hematoxylin binds to anionic tissue sites such as nuclear chromatin and stains the nucleus blue-black. Eosin, on the other hand, stains the cell cytoplasm and connective tissues different shades of red and pink.


45

Fig. 1. (A) Hematoxylin and eosin staining showings normal ectocervical epithelium. Arrow shows the normal basket-weave pattern of keratinocytes in the surface epithelium. The keratinocytes have regularly shaped nuclei and show maturation from the basal layer up to the epithelial surface. Note that there is no nuclear pleomorphism in the basal layer of the epithelium, which is usually one or two cells thick . (B) Cervical intraepithelial neoplasia (CIN) 1 with human papillomarvirus (HPV) changes. The short arrow shows a multinucleated keratinocyte and the long arrow indicates a koilocyte, both of which are pathognomonic of HPV infection. Koilocytes are characterized by enlarged and hyperchromatic nuclei, which are wrinkled in outline with perinuclear halo. Nuclear pleomorphism is apparent in the basal third of the epithelium.

3.3.1. Normal Ectocervix The epithelium of the transformation zone of the cervix consists of stratified squamous epithelium with a single basal cell layer with small, darkly stained cuboidal cells. The keratinocytes have regular-shaped nuclei and show maturation from the basal layer up to the epithelial surface, where the cells become progressively flattened until they are shed from the surface (Fig. 1A).

3.3.2. HPV-Associated Changes Koilocytes are virus-infected epithelial cells with enlarged irregular nuclei surrounded by clear cytoplasm and are indicative of productive HPV infection. Koliocytic cells have enlarged, wrinkled nuclei surrounded by perinuclear halo (Fig. 1B). However, HPV infection can be present in the ectocervical epithelium in the absence of koilocytes. Multinucleated epithelial cells in association with parakeratosis are termed HPV-like features.

46

Babawale et al.

Fig. 2. (A) Cervical intraepithelial neoplasia (CIN) 2. Note that there is nuclear abnormality in the full whole thickness of the epithelium, which is more pronounced in the lower two-thirds of the epithelium (arrow). (B) Cervical intraepithelial neoplasia (CIN) 3. Note lack of maturation of the keratinocytes throughout the full thickness of the epithelium. Arrow points to an abnormal mitotic figure high up in the epithelium. Both nuclear and cytological atypia of the keratinocytes up to the surface of the epithelium are seen.

3.3.3. CIN 1 (LGSIL) The cells in the lower third of the epithelium are hyperchromatic, showing a mild degree of pleomorphism and increased mitotic activity. Keratinocyte maturation is still seen in the upper two-thirds of the epithelium, and HPVassociated changes may be present (Fig. 1B).

3.3.4. CIN 2 (HGSIL) When basal cell proliferation extends up to two-thirds of the epithelial thickness, CIN2 is diagnosed. Moderate pleomorphism and abnormal mitoses are seen up to the middle third of the epithelial thickness, while keratinocyte maturation is still seen in the upper third (Fig. 2A).

3.3.5. CIN 3 and Carcinoma In Situ (HGSIL) Atypical cells extend into the upper third of the epithelium. Mitotic figures are common and are seen throughout the epithelium. Little cytoplasmic maturation may be still seen in the upper third of the epithelium in CIN3. However, no cytoplasmic maturation is seen in carcinoma in situ cases (Fig. 2B).

3.3.6. Invasive Cancer When irregular clumps of squamous cells are seen invading the stroma of the cervix, the case is then diagnosed as invasive cervical squamous cell carcinoma (SCC). The cells also show some atypical features, such as nuclear pleo-


47

Fig. 3. Invasive cervical squamous cell carcinoma. Arrow indicates a squamoid nest that has invaded the cervical stroma.

morphism, hyperchromasia, increased nuclear-cytoplasmic ratio, and frequent abnormal mitoses (Fig. 3).

3.3.7. HPV Typing Result Typical results of an HPV typing experiment are shown in Table 1. The analysis of 40 samples showed that nearly all the cases were PCR positive for HR-HPV and for HR-HPV16/E7 gene (see Table 1, showing representative 16/40 samples). Both HR-HPV viral load and HR-HPV16 E7 gene proportionally increased with increasing grade of cervical pre-cancer, and a highly significant association (p = 0.000) was found between the two. 4. Notes 1. Tissue sections are embedded in paraffin wax, which has to be removed prior to an experimental procedure by dipping the sections in xylene. Tissue sections are then clipped in ethanol to remove the traces of xylene and also to allow gradual rehydration of the tissue sections. 2. Differentiation is performed by dipping slides in 1% acid alcohol followed by “blueing” in water to ensure optimal staining. 3. Eosin Y (eosin yellowish, eosin water soluble) C.I. No 45380 (C.I. Acid Red 87) is the most widely used eosin. As a cytoplasmic stain, it is usually used as a 0.5 or 1.0% solution in distilled water, with a crystal of thymol added to inhibit fugal growth.

48

Babawale et al.

4. Differentiation of the eosin, by dipping slides in water and by dehydrating slides through ascending grades of ethanol, ensures that adequate intensity of the stain is maintained. 5. Dehydration through ascending grades of ethanol, followed by xylene, ensures that all traces of water are removed from the tissue sections. 6. Mounting of slides is carried out using distyrene (a polysterene), a plasticizer (tricresyl phosphate), and xylene. This mixture is known as DPX and is used as a mounting medium by placing a drop of DPX on cover slip and inverting the stained slide over the cover slip. 7. During DNA extraction from fixed tissues, xylene is added to a 30-µm section to remove paraffin wax. Ethanol is then added to remove the traces of xylene. 8. QuantiTect SYBR Green PCR master mix kit (purchased from Qiagen). This kit contains all the components for a successful real-time PCR. It contains optimized amounts of Taq DNA polymerase, a reference dye called ROX, a reporter dye SYBR Green I, nucleotides (dNTPs), and buffer. Qiagen HotStar DNA polymerase (included in the kit) prevents the production of nonspecific products at room temperature. 9. SYBR Green I detects all double-stranded DNA, and it is critical to determine whether there are any primer-dimers or nonspecific products being generated in the PCR. These appear as small peaks around low temperature (about 60–70°C). Using dissociation curve analysis for these sets of primers, no primer-dimers were detected.

Acknowledgments The authors wish to thank Ms. Anne Kane for excellent photo-imaging assistance. References 1. Zur Hausen, H. (1996) Papillomavirus infections-a major cause of human cancers. Biochem. Biophys. Acta 1288, F55–F78. 2. Jenkins, D. (2001) Diagnosing human papillomaviruses: recent advances. Curr. Opin. Infect. Dis. 14, 53–62. 3. Klaes, R., Friedrich, T., Spitkovsky, R., et al. (2001) Overexpression of p16 as a specific marker for dysplastic and neoplastic epithelial cells of the cervix uteri. Int. J. Cancer 92, 276–284. 4. Bancroft, J. D. and Stevens, A. (1996) Theory and Practice of Histological Techniques, 4th ed, Churchill Livingston, London, 99–110. 5. Jacobs, M., Sniders, P. J. F., van den Brule, A. J. C., Helmerhorst, T., Meijers, C. J. M., and Walboomers, J. M. M. (1997) A general primer GP5+/GP6+ mediated PCR-enzyme immunoassay method for rapid detection of 14 high-risk and 6 lowrisk human papillomavirus genotypes in cervical scrapings. J. Clin. Microbiol. 35, 791–795.

Detection of Papillomavirus Proteins and DNA

49

5 Detection of Papillomavirus Proteins and DNA in Paraffin-Embedded Tissue Sections Woei Ling Peh and John Doorbar Summary The key events during the papillomavirus life cycle can be mapped in infected tissue samples by antibody detection and in situ hybridization. The ease of immuno-detection varies for different proteins and is dependent on antigen availability. Epitope exposure is sometimes necessary, because the antigen may become masked after formalin fixation and paraffin embedding of the infected tissue. Visualization of both nucleic acid and protein targets can be done simultaneously by combining in situ hybridization and immuno-detection methods.

1. Introduction The most direct way to visualize the life-cycle events of papillomaviruses in an infected tissue biopsy is by immunodetection and in situ hybridization. In order to detect target proteins and viral DNA in cells, specific antibodies and DNA probes must first be made. Immunodetection of viral proteins using specific antibodies allows mapping of the protein expression patterns and sequence of events as they occur during an infection. Although a good antibody is critical for immunodetection experiments, the preparation and fixation of the tissue sample are also important to get successful staining results. Recent improvements to immunodetection techniques and reagents have facilitated high-resolution studies of proteins in vivo in formalin-fixed tissues. Even some proteins that in the past were difficult to detect using antibodies, can now be detected following antigen-retrieval treatments and signal amplification. This chapter describes some of the techniques and reagents that are commonly used for immunodetection studies on formalin-fixed, paraffin-embedded tissue sections. In addition, detection of viral DNA by in situ hybridization, and an outline for doing double staining experiments (protein–protein or protein–DNA), are also described. From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

49

50

Peh and Doorbar

2. Materials 2.1. Immunodetection of Proteins 1. Xylene (BDH Laboratories Supplies). 2. Absolute ethanol. 3. Phosphate-buffered saline (PBS): 142 mM NaCl, 2.7 mM KCl, 4 mM Na2HPO4, 1.8 mM KH2PO4. Adjust pH to 7.2 with HCl. 4. Hydrogen peroxide: 3% hydrogen peroxide made up in PBS. 5. Antigen-retrieval buffer (10 mM citrate buffer, pH 6.0): approx 43 mL of 0.1 M sodium citrate solution, 7 mL of 0.1 M citric acid solution in 500 mL buffer. 6. Trypsin solution: 0.1% (w/v) trypsin, 0.1% (w/v) calcium chloride, 20 mM TrisHCl solution (pH 7.8). 7. Blocking solution: 10% (v/v) normal goat serum made up in PBS. 8. Wash buffer: 0.05% (v/v) Tween-20™ (Sigma-Aldrich Company, Ltd.) in PBS. 9. Tris-buffered saline (TBS): 142 mM NaCl, 2.7 mM KCl, 25 mM Tris-HCl. Adjust pH to 8 with HCl. 10. Specific antibodies against target proteins (see Note 1). 11. Fluorophore-labeled or enzyme-linked species-specific secondary antibodies. 12. 3',3'-Diaminobezidine tetrahydrochloride tablet (DAB; Sigma-Aldrich Company, Ltd.). 13. Sigma Fast™ Fast Red TR/Naphthol AS-MX Tablet Sets (Sigma-Aldrich Company, Ltd.). 14. Nuclei counterstains: 4', 6-diamidino-2-phenylindole dihydrochloride (DAPI; Sigma-Aldrich Company, Ltd.), used at 0.5 µg/mL, which gives a blue signal. Propidium iodide (Sigma-Aldrich Company, Ltd.) used at 0.5 µg/mL, which gives a red signal. 15. Cellular counterstain: Harris’s hematoxylin (BDH Laboratories Supplies), 0.5% glacial acetic acid/99.5% ethanol. 16. Mounting media: Citifluor (Agar Scientific, UK), DPX (BDH Laboratories Supplies). 17. Coverslips (BDH Laboratories Supplies). 18. Wilson jar or slide troughs and holders (Fisher Scientific). 19. Microwave. 20. Pressure cooker. 21. Humidified box. 22. ImmEdge pen (Vector Laboratories Incorporated, Burlingame, CA). 23. Microscope equipment, imaging instruments, and software for viewing and acquiring fluorescent images. 24. Light microscope.

2.2. Additional Materials for DNA Fluorescent In Situ Hybridization (FISH) 1. Proteinase K solution: 50 µg/mL proteinase K in PBS. 2. 20X SSC: 88.2 g tri-sodium citrate, 175.3 g NaCl per liter. Adjust pH to 7.8 with sodium hydroxide (see Note 2).


51

3. Hybridization buffer: 50% (v/v) deionized formamide, 1X Denhardt’s solution (Sigma-Aldrich Company, Ltd.), 5% (w/v) dextran sulphate, 200 µg/mL salmon sperm DNA, 4X SSC. 4. DIG-labeled DNA probes: Digoxigenin (DIG)-labeling kit (Roche Diagnostics), DNA template. 5. Anti-DIG horseradish peroxidase (HRP)-conjugated antibodies (Roche Diagnostics). 6. Formamide wash buffer: 50% (v/v) formamide, 2X SSC, 0.05% (v/v) Tween-20. 7. TE buffer: 10 mM Tris-HCl (pH 8.0), 1 mM ethylenediamine tetraacetic acid (EDTA). 8. Hot block.

2.3. Signal-Amplification Systems 1. ABC amplification: ABC Reagents (DAKO, Ltd.) and biotin-labeled speciesspecific secondary antibodies. 2. TSA™ (tyramide signal amplification) fluorescence systems (NEN Life Science Products, Inc.).

3. Methods This section describes the methods by which (1) viral proteins can be detected using specific antibodies, (2) viral DNA can be detected using DIGlabeled DNA probes by in situ hybridization, and (3) protein–protein or protein–DNA double stains can be carried out on the same tissue section. In addition, the use of signal-amplification systems to improve staining results is discussed.

3.1. Detection of Proteins in Paraffin-Embedded Tissue Sections The use of specific antibodies for the detection of papillomavirus proteins in infected tissue sections has been shown for E2, E4, E7, L1, and L2 proteins (1–5). However, the ability to detect and visualize these proteins in vivo varies considerably. In general, E4, L1, and L2 proteins can be readily detected using specific antibodies, whereas the detection of E2 and E7 requires additional antigen-retrieval steps and/or signal-amplification steps.

3.1.1. Preparation of Tissue Sections for Immunostaining 1. Prepare tissue sections (5 µm thickness; see Note 3) from paraffin-embedded tissue blocks (see Chapter 3) onto precoated microscope slides (see Note 4). 2. Deparaffinize and rehydrate the tissue sections: soak the slides twice in xylene (1 × 10 min, 1 × 5 min), twice in 100% ethanol (2 min each), followed by a series of graded ethanols (80%, 50%, 30%) for 2 min each, and finally distilled water (at least 5 min).

52

Peh and Doorbar

3.1.2. Antigen-Retrieval Treatments The two major ways in which epitopes may be retrieved are by (a) heat denaturation (microwave or pressure cooking) and (b) proteolytic treatment (see Note 5). 3.1.2.1. MICROWAVE TREATMENT 1. Place slides in a glass slide holder and presoak for 5 min in 500 mL of antigenretrieval buffer (see Note 6) in a 1-L plastic beaker (see Note 7). 2. Cook (at 650 W) for 12–15 min (3 × 5 min, with 1 min intervals), then cool for 15–20 min. 3. Wash in PBS for 5 min.

3.1.2.2. PRESSURE COOKING 1. Make up enough citrate buffer to fill one-quarter of the pressure cooker (the buffer level must be able to completely cover the slides in a slide holder). 2. Bring the buffer to a boil in the pressure cooker and carefully place the presoaked slides (5 min in antigen-retrieval buffer) into the boiling buffer. 3. Cook the slides at full pressure for 2–10 min (length of treatment should be optimized for each antigen). 4. Cool the cooker in a basin of cold tap water. (At this point, it is very important to wait until the pressure is completely released from the cooker before it is safe to open the lid.) 5. Leave the slides to cool for 20 min before washing in PBS for 5 min.

3.1.2.3. PROTEOLYTIC TREATMENT 1. Digest tissue sections with pre-warmed (37°C) trypsin solution for 30 s to 20 min (depending on the antigen and tissue) in a humidified box (see Note 8). 2. Wash the slides for 2 × 5 min in PBS.

3.1.3. Immunodetection 1. Circle the perimeter of the tissue section using an ImmEdge pen. 2. Block with 10% normal goat serum/PBS (see Note 9) at room temperature for 1 h in a humidified box. 3. Dilute the antigen-specific primary antibody in 5% normal goat serum/PBS, add the antibody onto the section, and incubate at room temperature for at least 1 h in a humidified box (see Note 10). 4. Wash the section for 3 × 5 min in wash buffer at room temperature with shaking. 5. Dilute the fluorophore-labeled species-specific secondary antibodies (see Note 11) in either PBS or in 5% normal goat serum/PBS. Add a nuclear counterstain to show the organization of the epidermis. 6. Incubate the tissue section with the secondary antibody and nuclear counterstain mix at room temperature for 30–60 min in a dark humidified box. 7. Wash the section for 3 × 5 min in wash buffer at room temperature with shaking.


53

Alternatively, enzyme-linked HRP or alkaline phosphatase [AP]) speciesspecific secondary antibodies (see Note 12) can be used in step 5. 1. For HRP antibodies, incubate the tissue section with 3% hydrogen peroxide at room temperature for 15 min to quench endogenous peroxidase activity. 2. Dilute the enzyme-linked species-specific secondary antibodies in either PBS or in 5% normal goat serum/PBS (see Note 13 for AP-linked antibodies). For fluorescence studies, add an appropriate nuclear counterstain at this step. 3. Incubate at room temperature for 30–60 min in a dark, humidified box. 4. Wash the section for 3 × 5 min in wash buffer at room temperature with shaking.

3.1.4. Visualization 3.1.4.1. IMMUNOFLUORESCENCE 1. Wash the section briefly in PBS and mount with Citifluor. 2. View results by fluorescent microscopy.

3.1.4.2. ENZYME-SUBSTRATE DETECTION 1. Prepare substrate solutions according to the manufacturer’s instructions. Substrates for HRP activity: tyramide-fluorophore (fluorescent) and DAB (nonfluorescent). Substrate for AP activity: Fast Red (see Note 13). 2. Add the substrate solution onto the section. 3. For tyramide-fluorophore substrates, incubate for 8–10 min, then wash in PBS for 5 min. Mount onto slides with citifluor. 4. For DAB and Fast Red substrates, monitor color development on the tissue section under a light microscope to prevent the accumulation of background staining. Stop the reaction by washing 2 × 5 min in distilled water. 5. Counterstain for nonfluorescent staining. Add a few drops of Harris’s hematoxylin to the section for approx 5 min. Wash gently in distilled water. Dip in glacial acetic acid/ethanol for 10 s. Wash in distilled water. Air-dry and mount with DPX.

3.2. Detection of Viral DNA Using DIG-Labeled DNA Probes This protocol is optimized for FISH, with the use of random-primed, DIGlabeled DNA probe, and visualized with a tyramide-fluorophore substrate.

3.2.1. Synthesis of DIG-Labeled Probes 1. Linearize the viral genome DNA by enzyme digestion, gel purify, and use this as template for making the DIG-labeled DNA probe. 2. Prepare the labeling reaction using the DIG-labeling kit according to the manufacturer’s instructions. 3. Incubate the reaction in a 37°C water bath overnight. 4. Purify the DIG-labeled probe according to the manufacturer’s instructions and store it at –20°C. (Optional: labeling efficiency may be tested by dot-blotting.)

54

Peh and Doorbar

3.2.2. In Situ Hybridization 1. Deparaffinize and rehydrate the tissue sections as described under Subheading 3.1.1. 2. Incubate the tissue section in 3% hydrogen peroxide for 15 min at room temperature (see Subheading 3.1.3.). 3. Wash the section for 5 min in PBS. 4. Digest the section with a freshly made-up proteinase K (50 µg/mL) solution for 20 min at 37°C in a humidified box. 5. Wash the section for 2 × 5 min in PBS, then allow to air-dry. 6. Dilute the DIG-labeled DNA probe (1:25) in hybridization buffer. 7. Add 10–50 µL of the diluted probe mix (depending on the size of the tissue section) to the tissue section. 8. Carefully lower a coverslip onto the tissue section, and ensure that no air bubbles are trapped under the coverslip, and the entire tissue section is covered with the probe mix. 9. Heat the section on a hot block at 95°C for 5 min, then cool on ice for at least 1 min. 10. Incubate the section overnight at 37°C in a humidified box.

3.2.3. Detection and Visualization 1. Loosen the coverslip by shaking the slide in 0.5X SSC. 2. Wash the section with prewarmed (42°C) solutions—formamide wash buffer for 2 × 5 min, then 2X SSC for 2 × 5 min. 3. Block with 10% normal goat serum/PBS for 1 h at room temperature in a humidified box. 4. Prepare HRP-conjugated anti-DIG antibodies and a nuclear counterstain in 5% normal goat serum/PBS. Incubate for 1 h at room temperature in a dark humidified box. 5. Wash the section for 3 × 5 min in wash buffer. 6. Prepare and use the tyramide-fluorophore substrate solution (see Subheading 3.1.4.) for the detection of HRP activity.

3.3. Protein–Protein or Protein–DNA Double Detection in a Tissue Section To perform co-localization studies (protein–protein or protein–DNA) on the same tissue section by immuno-fluorescence techniques, several factors have to be considered beforehand (see Note 14).

3.3.1. Immunodetection of Different Proteins 1. Prepare tissue section for immunodetection as described under Subheadings 3.1.1. and 3.1.2. 2. Block the section with 10% normal goat serum/PBS at room temperature for 1 h in a humidified box. 3. Dilute the two primary antibodies in 5% normal goat serum/PBS and add this onto the section for at least 1 h in a humidified box (see Note 10).


55

4. Wash the section for 3 × 5 min in wash buffer at room temperature with shaking. 5. Dilute fluorophore-labeled species-specific secondary antibodies in either PBS or in 5% normal goat serum/PBS. Add a nuclear counterstain to show the organization of the epidermis. 6. Incubate the section with the secondary antibody and nuclear counterstain mix at room temperature for 30–60 min in a dark humidified box. 7. Wash the section for 3 × 5 min in wash buffer at room temperature with shaking.

3.3.2. Combining FISH and Immunodetection 1. Prepare the tissue section for immunodetection as described under Subheadings 3.1.1. and 3.1.2. (see Note 15). 2. Process the section for ISH and hybridize at 37°C overnight as described under Subheading 3.2.2. 3. Loosen the coverslip by shaking the slide in 0.5X SSC. 4. Wash the section with prewarmed (42°C) solutions—formamide wash buffer (2 × 5 min), then 2X SSC (2 × 5min). 5. Block the section in 10% normal goat serum/PBS with blocking buffer for 1 h at room temperature in a humidified box. 6. Add the primary antibody mix onto the section for protein detection as described under Subheading 3.1.3. 7. Wash the section for 3 × 5 min in wash buffer. 8. Prepare a mixture containing the HRP-conjugated anti-DIG antibodies, a fluorophore-labeled, species-specific antibody (which recognizes the primary target antibody), and a nuclear counterstain in 5% normal goat serum/PBS. Add onto the section and incubate for 1 h at room temperature in a dark humidified box. 9. Wash the section for 3 × 5 min in wash buffer. 10. Prepare and use the tyramide-fluorophore substrate solution (see Subheading 3.1.4.) for the detection of HRP activity.

3.4. Signal-Amplification Systems Signal-amplification systems can significantly improve immunostaining results that are weak or difficult to obtain as a result of low antibody affinity or low protein expression levels (see Note 16).

3.4.1. ABC Reagents 1. Following primary antibody binding, detect the primary antibody with a biotinlinked, species-specific secondary antibody. Incubate the section with the biotinylated antibody at room temperature for 30–60 min in a humidified box. 2. Prepare the StreptABComplex/enzyme mixture according to the manufacturer’s instructions. 3. Wash the section for 3 × 5 min in wash buffer (see Note 10). 4. Add the StreptABComplex mixture onto the tissue section and incubate for 30 min at room temperature.

56

Peh and Doorbar

5. Wash the section for 3 × 5 min in wash buffer. 6. Prepare the appropriate substrate solutions for the detection of HRP or AP activities.

3.4.2. TSA Fluorescence Systems 1. Prepare the substrate according to the manufacturer’s instructions. 2. Add the substrate onto the tissue section and incubate for 8–10 min in a dark, humidified box. 3. Wash the section in PBS for 5 min. 4. Mount with citifluor.

4. Notes 1. Primary antibody binding affinity and specificity to the target antigen should be tested on Western blots and enzyme-linked immunosorbent assay (ELISA) prior to use on tissue sections. For some antibodies, antigen recognition on Western blots and ELISA does not guarantee antigen detection on tissue sections. Optimal dilutions for each antibody must be individually tested. This is particularly important for polyclonal antisera generated in animals, as some antibodies may be best used at 1:40 while others give clean staining only when used at 1:6000. A good working dilution to start with for polyclonal antibody is 1:250. Monoclonal antibody is usually used at a more dilute concentration (for example, 1:1000), while hybridoma supernatant is used at a higher concentration (for example, 1:100 or less). 2. A stock solution of 20X SSC can be made up, autoclaved, and stored at room temperature. Once made up, the hybridization buffer may be stored at –20°C for at least 3 mo. 3. When preparing tissue sections onto slides, it is useful to label each section in a serial order as they are being cut. If different staining experiments are carried out on sequentially cut tissue sections, the expression pattern of viral/host proteins (detected by immunostaining) in a particular region of interest (for instance an area where the late stage of the papillomavirus life cycle is supported) may be compared. Although comparison of different staining patterns can be done on adjacent sections, this method cannot be used for co-localization studies of different proteins. 4. Different kinds of coated microscope slides for the preparation of tissue sections are commercially available today. We have found that charge-coated slides (BDH Laboratories Supplies) are superior to poly-lysine-coated slides in withstanding harsh antigen exposure treatments. It is also useful to incubate the tissue sections at 37°C overnight prior to immunodetection, as this seems to reduce the chance of tissue detachment during staining experiments. 5. This step is optional and can be performed to expose antigen epitopes that may be masked by formalin fixation (6,7). Antigen retrieval before antibody detection is necessary for some proteins, but not for others. Combination of heat denaturation and proteolytic treatment is also used for some proteins, such as HPV-11 E1^E4


6.

7.

8. 9.

10.

11.

12.

57

(8,9). Heat denaturation is usually carried out first, followed by proteolytic digestion (10). The antigen-retrieval buffer may be modified for different antigens. Other buffers frequently used in our laboratory include 10 mM citric acid buffer (4) and 10 mM citrate buffer containing 1 mM EDTA (pH 8.0) (9). Antigen-retrieval buffers of higher pH (8.0–9.9), are recommended by some researchers and claimed to improve staining of some proteins (10,11). The beaker should be covered with paper towels to prevent excessive buffer loss during the microwave process. Paper towels should be secured with tape and perforated to allow gaseous escape. It is also useful to note the level of buffer in the beaker before the microwave treatment, as the buffer level is topped up during the cooling step. In our hands, trypsin digestion of approx 1–2 min is usually sufficient. It is crucial not to over-digest the section to prevent tissue damage and/or loss of epitope. 10% fetal bovine serum (not fetal calf serum, as this contains higher amounts of immunoglobulin [Ig]G), 3% bovine serum albumin, and 10% nonfat dry milk are also commonly used as blocking solutions for tissue staining (12). The optimal length and temperature of antibody binding may vary for different antibodies. Lower temperatures (such as 4°C) are preferred in some cases, as high temperatures may cause the loss of some epitopes (11). However, we have observed that some antibodies produced much cleaner staining results when used at 37°C instead of room temperature or 4°C. Sensitivity of detection may be increased with longer or overnight incubations. Fluorophore-labeled secondary antibodies are useful for doing high-resolution studies, but the main disadvantage is signal fading with increasing excitation radiation exposure (photo-bleaching), as well as with increasing time. The most commonly used fluorophores are fluorescein, fluorescein isothiocyanate (FITC), rhodamine, and Texas red. Antibodies labeled with these fluorophores are widely available commercially. However, we have found that the Alexa fluor dye series (Molecular Probes, Inc.) are superior to these traditional fluorophores. As the manufacturer has claimed, we find that the Alexa fluor dyes are more photostable and produce a more intense fluorescence signal. When working with fluorophores, it is advisable to perform antibody incubations in the dark and reduce the excitation radiation exposure time to minimize photo-bleaching. Fluorescent stained and mounted sections may be stored at 4°C for several months. Enzyme-based detection methods are versatile for protein localization studies in cells, since different fluorescent and non-fluorescent substrates are now widely available for the enzymes. Colored fluorophores have been developed for HRP activity, and we have found the tyramide-based fluorophores (see Subheading 3.4.2.) are reliable in producing relatively clean staining results. On the other hand, non-fluorescent substrates such as DAB (for HRP activity) and Fast Red (for AP activity) are also commonly used for immuno-histochemistry. Nonfluorescent substrates are cheap, and ideal for producing permanent results that can be stored for record keeping. In addition, special microscope equipment

58

13.

14.

15. 16.

Peh and Doorbar and imaging instruments and software are not required for viewing nonfluorescent staining. It should be noted that all phosphate-based buffers (for antibody dilution and washing) should be replaced with Tris-based buffers (TBS) when using AP. Fast Red can be used as a fluorescent substrate for AP, and can be viewed under a fluorescent microscope using a Texas Red (or equivalent) microscope filter. However, it is not advisable to use Fast Red as a fluorescent substrate in double immuno-staining experiments, as it is an opaque substrate and can cause masking of other fluorophores in co-localization studies. Antibody selection. When performing double immunostaining experiments, it is important to ensure that the origins of both primary antibodies are derived from distantly related animal species—for example, rabbit and mouse. The use of primary antibodies from closely related animal species, such as rat and mouse, is not recommended for co-localization studies, although this is not impossible. In this case, staining of the individual proteins can be done on adjacent tissue sections so that the separate staining patterns within a specific region of the tissue may be compared and used to verify the double immunostain results. This will help to rule out the possibility of cross-reactivity of the secondary antibodies that may contribute to false-positive staining results in double immunostained sections. Alternatively, the use of direct fluorophore-conjugated primary antibodies eliminates the possibility of secondary antibody cross-reactivity. Otherwise, biotin- or DIG-labeled primary antibodies may also be useful in multiple protein immunostain experiments. Instead of species-specific antibodies, streptavidin conjugates or anti-DIG antibodies may be used. Microscopy. Fluorescent microscopy is an area where potential false-positive results may arise. False-positive results may be obtained when bleed-through from the microscope color filters occurs. For example, red fluorescence may be seen through a green filter at the green fluorescence wavelength setting, even though the red fluorescence should not be excited at that specific wavelength. Color filter bleed-through can be avoided by using a color filter with a narrow spectrum range that is optimally designed for your choice of fluorophore. Another common contributor to color bleed-through in multiple-color fluorescence studies is any particular fluorescent color being too bright (signal saturation). This can be prevented by optimizing the concentration of the primary and secondary antibodies. We have found that antigen retrieval is sometimes not necessary for certain antigens when the tissue section is prepared for ISH. Signal amplification systems: The ABC reagents work by amplifying the signal of one biotin-labeled secondary antibody into a large multimeric enzyme complex (StreptABComplex/enzyme) that can be detected by appropriate enzyme substrates. The TSA fluorescence systems are available in five different tyramidefluorophore substrates (fluorescein, tetramethyl-rhodamine, coumarin, cyanine 3, and cyanine 5) for the detection of HRP activity in tissue sections or cells. The tyramide-fluorophore substrates can be used in indirect immunodetection of pro-


59

teins (see Subheading 3.1.3.), or in conjunction with ABC reagents (see Subheading 3.4.1.). We have found that tyramide fluorophores are particularly valuable in enhancing fluorescent signals with minimal loss in resolution.

References 1. Maitland, N. J., Conway, S., Wilkinson, N. S., et al. (1998) Expression patterns of the human papillomavirus type 16 transcription factor E2 in low- and high-grade cervical intraepithelial neoplasia. J. Pathol. 186(3), 275–280. 2. Penrose, K. J. and McBride, A. A. (2000) Proteasome-mediated degradation of the papillomavirus E2-TA protein is regulated by phosphorylation and can modulate viral genome copy number. J. Virol. 74(13), 6031–6038. 3. Du, J., Chen, G. G., Vlantis, A. C., et al. (2003) The nuclear localization of NFkappaB and p53 is positively correlated with HPV16 E7 level in laryngeal squamous cell carcinoma. J. Histochem. Cytochem. 51(4), 533–539. 4. Middleton, K., Peh, W., Southern, S., et al. (2003) Organization of human papillomavirus productive cycle during neoplastic progression provides a basis for selection of diagnostic markers. J. Virol. 77(19), 10,186–10,201. 5. Doorbar, J., Foo, C., Coleman, N., et al. (1997) Characterization of events during the late stages of HPV16 infection in vivo using high-affinity synthetic Fabs to E4. Virology 238(1), 40–52. 6. Shi, S. R., Key, M. E., and Kalra, K. L. (1991) Antigen retrieval in formalinfixed, paraffin-embedded tissues: an enhancement method for immunohistochemical staining based on microwave oven heating of tissue sections. J. Histochem. Cytochem. 39(6), 741–748. 7. Cattoretti, G., Becker, M. H., Key, G., et al. (1993) Antigen unmasking on formalin-fixed, paraffin-embedded tissue sections. J. Pathol. 171(2), 83–98. 8. Bryan, J. T. and Brown, D. R. (2000) Association of the human papillomavirus type 11 E1()E4 protein with cornified cell envelopes derived from infected genital epithelium. Virology 277(2), 262–269. 9. Peh, W. L., Middleton, K., Christensen, N., et al. (2002) Life cycle heterogeneity in animal models of human papillomavirus-associated disease. J. Virol. 76(20), 10,401–10,416. 10. Grafisk, K. (1997) A Guide to Demasking of Antigen on Formalin-Fixed, Paraffin-Embedded Tissue 2nd ed. DAKO, Copenhagen, Denmark. 11. Harlow, E. and Lane, D. (1999) Using Antibodies, A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. 12. Harlow, E. and Lane, D. (1998) Antibodies, A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.

HPV Gene Expression Using Real-Time PCR

61

6 Detection and Quantitation of HPV Gene Expression Using Real-Time PCR Rashmi Seth, John Rippin, Li Guo, and David Jenkins Summary Human papillomaviruses (HPVs) are known to be etiological agents of cervical cancer and have been found in 99.7% of women with high-grade (HG) cervical intraepithelial neoplasia (CIN) precancer. Testing of high-risk HPV (HR-HPV) has been proposed as a way of improving cervical screening, especially for women with low-grade (LG) Papanicolaou (Pap) smears. In this chapter, real-time quantitative polymerase chain reaction (PCR) methods that can be used to investigate the expression of HPV 16 early genes in HG or LG precancer are demonstrated. Detecting the expression of early HPV genes in conjunction with the Pap smear may improve the specificity of identifying LG precancers that are associated with high risk of progression.

1. Introduction High-risk human papillomaviruses (HPVs) are associated with cervical intraepithelial neoplasia (CIN) and cervical carcinoma, and HPV DNA has been found to be present in more than 99.7% of these cancers (1). The highrisk group so far includes HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 54, 56, 58, 59, 66, 68, 70, and 72. The infecting HPV type, the viral load, and the integration state of the HPV genome are known to have profound implications for patient prognosis (2–4). In countries such as the United Kingdom and the United States, national cervical screening programs have reduced the incidence of cervical cancer (5). However, 50% of invasive cervical cancers arise in women screened with existing cytological methodologies (6). HPV detection and typing techniques have been proposed as an adjunct to, or a replacement for, the current cytological screening regime, and the success of such strategies will depend on the development of rapid, sensitive, and specific HPV detection methods applicable in the clinical setting. From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

61

62

Seth et al.

Many approaches to the detection and typing of HPV are available, all of which have advantages and disadvantages, and have been reviewed elsewhere (7). Various polymerase chain reaction (PCR)-based tests have been developed to detect HPV DNA using primers that amplify a region of the major viral capsid L1 gene that is highly conserved. They include the consensus primers MY09/MY11 (8), PGMY09/11 (9), and GP5+/GP6+ (10). All assays require post-PCR processing to identify positivity and individual genotypes. For individual genotyping, a pool of type-specific primers for use in PCR have been developed, followed by reverse line blot techniques or Southern blotting (11–13). With the availability of quantitative real-time PCR instruments and different chemistries, new assays for HPV have already been reported. Swan and others used Taqman probes to detect and quantitate HPV genotypes (14). Others include using self-probing amplicons known as Scorpions (15). Molecular beacon-labeled primers (16) have also been developed and used by Jordens and others (17) to detect and genotype HPV. The main disadvantages of these techniques are the cost of synthesis of dye-labeled primers and probes, and unavailability of a continuous and reliable supply. The “gold standard” for HPV genotyping is sequencing, in which HPV DNA is amplified and the product is sequenced. Although this technique gives the most conclusive genotype information, it is also the most labor-intensive and costly. There is a need for a rapid, cost-effective test that can distinguish between the different HPV genotypes on the basis of their nucleotide sequences, and two such methods have been reported using real-time PCR technique (18,19). This approach combines PCR amplification with post-PCR amplicon meltcurve analysis, and it involves using the fluorogenic dye SYBR® Green I. Viral load in each sample can be quantified and the HPV genotype interrogated by a post-PCR melt-curve analysis. Our studies as well as those of others have shown that HPV 16 is the most frequently detected genotype. In this chapter we demonstrate a method for the detection and analysis of the HPV 16 early genes E2, E4, and E6 in RNA swabs from clinical specimens using quantitative real-time PCR and SYBR Green I dye. 2. Materials 1. 2. 3. 4.

Real-time PCR instrument (Mx4000, Stratagene, UK). Primers synthesized by UK-OSWEL, UK. cDNA from clinical specimens. QuantiTect SYBR Green Master Mix (Qiagen, UK). The kit contains a vial of master mix, which is a premixed solution containing HotStarTaq DNA polymerase, PCR buffer, dNTPs, SYBR Green I dye, a reference dye ROX, and a vial of PCR-grade water.


63

Table 1 Summary of Primer Sequences for Human Papillomavirus (HIV) 16 E-Genes and Corresponding PCR Product Size PCR type

Sequence

Site in HPV genome

Product

HPV16 E2

Forward 5'-gccaacactggctgtatcaa-3' Reverse 5'-catcctgttggtgcagttaaa-3' Forward 5'-tccaatgccatgtagacgac-3' Reverse 5'-gctcacacaaaggacggatt-3' Forward 5'-gcataaatcccgaaaagcaa-3' Reverse 5'-agcgacccagaaagttacca-3'

2932–2951 3061–3081 3700–3719 3573–3592 287–306 123–142

149 bp

HPV16 E4 HPV16 E6

146 bp 134 bp

5. Standard HPV 16 viral DNA control, 50 ng/µL (purchased from Advanced Biotechnologies Inc. UK). This DNA is purified from CaSki cells, which are reported to contain approx 600 copies of integrated HPV-16 DNA per cell.

3. Methods The methods described below outline (1) primer pair design for HPV 16 genes E2, E4, and E6, (2) optimization of primers for use in real-time PCRs, and (3) how to interpret the data obtained from clinical specimens.

3.1. Designs and Development of Primers Primers can be designed in-house, using the Web-based primer design program Primer 3, available on the Internet. (http://www-genome.wi.mit.edu/cgibin/primer/primer3_www.cgi). The software allows the lengths of the primers and annealing temperatures for the PCR to be customized and also gives the size of the target PCR product. Primer sequences once identified can then be synthesized commercially (see Note 1). In Table 1, the nucleotide sequences are listed, together with the expected product size and the position in the HPV 16 genome.

3.2. HPV 16 E-Gene Real-Time PCRs Before commencing any real-time PCRs using SYBR Green I, it is vital to optimize the primer concentrations and thermal cycling conditions to eliminate nonspecific product formation. Once these have been determined, the assays can be validated in terms of analytical specificity and sensitivity before applying them.

3.2.1. Optimization of Primers We use a kit (QuantiTect SYBR Green PCR master mix) for real time PCRs (see Notes 2 and 3). Each PCR uses a single set of primers in a single reaction

64

Seth et al.

Fig. 1. An example of a thermal profile for human papillomavirus (HPV) 16 E4 gene polymerase chain reaction.

mixture. A checkerboard experiment (not shown) is set up with varying forward primer dilutions across the horizontal axis and the reverse primer dilutions along the vertical axis (varying from 0.1 pmol to 500 pmol/tube in 1 µL volume). The following protocol is used. 1. 12.5 µL of master mix (provided in the kit) is added to each tube. 2. Add 1 µL of each primer at varying dilutions. 3. DNA template (2 µL HPV16 DNA, 2 ng/tube) is added next to all the tubes except those that are to be kept as negative controls. 4. Finally, 8.5 µL of RNAse-free, PCR-grade water (provided in the kit) is added to obtain a final volume of 25 µL in each tube. 5. Perform forty cycles of PCR amplification using a general thermal profile shown in Fig. 1. 6. Interpretation of results (see Note 4).

Ct (cycle threshold) values for each dilution are computed from the amplification plots using the built-in software, and are shown in Table 2. The final primer dilutions for E2 gene PCR were determined as 35 pmol for both forward and reverse primer. For the E4 and E6 gene PCRs, the primer dilutions were set at 25 pmol for both forward and reverse primers.

Table Gene 2 HPV Expression Using Real-Time PCR Cycle Threshold (Ct) Values for Each Primer Combination for E2, E4, and E6 HPV 16 Assays Primers

65

Forward primer concentration

Reverse primer concentration

䉴

100 pmol, 100 pmol E2 Ct = 15.86 E4 Ct = 13.60 E6 Ct = 13.52

100 pmol 35 pmol E2 Ct = 16.22 E4 Ct = 13.39 E6 Ct = 13.55








3.2.2. Optimizing Annealing Temperature (Thermal Profiles) The best annealing temperatures for real-time PCRs should be investigated (see Note 5). The following protocol can be used to optimize the annealing temperature for any real-time PCR. 1. Set up master mixes for each PCR as previously, but using the optimized primer concentrations (for E2 gene, 35 pmol; for E4 and E6, 25 pmol) and HPV16 DNA standard (2 ng/tube). 2. Set up tubes containing cloned HPV types (HPV 6, 11, 16, 18, 31, 33, 35, 39, 45, 51, 56, 58, 59, 66, and 68) at approx 100 pg/tube to identify the effect of each temperature on primer binding specificity. 3. Different annealing temperatures (61, 59, 54, and 52°C) must be tested (see Note 5). For HPV16 E2, the best annealing temperature was found to be 54°C, whereas for HPV E4 and E6 it was 59°C.

3.2.3. Specificity (Cross-Reactivity) Analysis Once the optimal thermal profile has been identified, the specificity of each gene assay must be investigated by setting up cross-reactivity studies. We use DNA of known HPV types (6, 11, 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 66, and 68) (see Note 6) at concentrations of 100 pg/tube. 1. Set the PCRs up as before in a final volume of 25 µL per tube, which contains 12.5 µL of the QuantiTect master mix, 1 µL of the forward primer, 1 µL of the reverse primer, 2 µL of the template, and 8.5 µL of PCR-grade water.

66

Seth et al.

Table 3 Effect of Assay Sensitivity and Cross-Reactivity When the Annealing Temperature Is Altered Temperature 52°C Annealing temperature

55°C annealing temperature



HPV E2

HPV E4

HPV E6

Cross reacts with 18, 31, 35, 39, 45, 51, 66, 6, 11

Cross reacts with 18, 31, 33, 35, 39, 45, 51, 56, 58, 59, 66, 6, 11 HPV 16 sensitivity = 0.0002 ng Cross reacts with 18, 31, 33, 35, 39, 45, 51, 59, 66, 6, 11 HPV 16 sensitivity = 0.0002 ng Cross reacts with 18, 31, 33, 66, 6 HPV 16 sensitivity = 0.0002 ng Cross reacts with 18, 31, 66, 6 HPV 16 sensitivity = 0.02 ng

Cross reacts with 18, 31, 33, 35, 39, 45, 51, 56, 58, 59, 66, 6, 11 HPV 16 sensitivity = 0.0002 ng Cross reacts with 18, 31, 33, 35, 39, 45, 51, 59, 66, 6, 11 HPV 16 sensitivity = 0.0002 ng Cross reacts with 18, 31, 33, 39 HPV 16 sensitivity = 0.0002 ng Cross reacts with 18, 33, 39 HPV 16 sensitivity = 0.02 ng

HPV 16 Sensitivity = 0.0002 ng Cross reacts with 18, 31, 33

HPV 16 sensitivity = 0.0002 ng Cross reacts with 18 HPV 16 sensitivity = 0.02 ng Cross reacts with 6 HPV 16 sensitivity = 0.2 ng

2. Carry out the PCR using the optimal thermal profile identified in the previous experiment.

In Table 3, the types of results that can be expected are shown. HPV 18, 31, and 33 cross-react with HPV 16 E2 gene; HPV 18, 31, 33, 66, and 6 cross-react with HPV16 E4 gene; and HPV 18, 31, 33, and 39 cross-react with HPV 16 E6 gene at a low rate of 0.2%. This is because the nucleotide sequences in the HPV 16 E-genes are very similar to those of HPV18, 31, 33, and 39 E-genes.

3.2.4. Reproducibility of the HPV PCRS To test for intra- and inter-assay variations, the following method can be used: 1. Dilute HPV 16 DNA standard in PCR water to a concentration of 0.25 ng/µL in a volume of 200 µL. 2. Using the optimized PCR reaction mixtures and thermal profiles for E2, E4, and E6 PCRs, set the PCR assays as before. 3. Test the sample (0.5 ng/tube, 2 µL) 20 times in a single run (to obtain intra-assay variation) and 10 times in consecutive runs (to obtain inter-assay variation).


67

4. Calculate the mean Ct values, the standard deviations, and the percentage coefficient of variation (see Note 7).

3.2.5. Standard (Calibration) Graphs to Analyze Assay Sensitivity Analytical sensitivity of a PCR assay is defined as the least amount of specific PCR product that can be detected with confidence. To determine sensitivity of the E-gene PCRs, the following procedure can be followed: 1. Make a 10-fold serial dilution using HPV16 DNA standard to cover the range from 2 ng/tube to 0.00002 ng/tube in PCR-grade water. 2. Use 96-well format to set up the PCRs. 3. Pipet standards (2 µL) in duplicate into the tubes. 4. Also, set up 10 wells as negative controls (no template controls, NTC) to monitor contamination. 5. Perform PCR reactions as before in 25µL final volume using thermal profiles for HPV 16 E2, E4, E6 followed by dissociation curve analysis.

Fluorescence data collected during the PCR is fed into the software, which constructs a standard curve on the bases of the Ct value of each well. Quantity of the PCR product for each of the PCR reactions is then calculated from this information. A text report is obtained, which tabulates the quantity of the PCR product in ng/tube. Two major factors are considered when assessing sensitivity of the PCR. They are (1) the analytical efficiency and specificity of the reaction, preventing potential primer-dimer product formation and (2) maximum discrimination between negative controls and the least detectable specific product (see Note 8). In our hands, the assay sensitivity was found to be 0.0002 ng/tube for the early genes E2, E4, and E6.

3.3. Clinical Application of HPV PCRs To analyze gene expression, total RNA (cDNA) is needed from clinical specimen. Total RNA used in our study was obtained during the TOMBOLA clinical trial—a trial of management of borderline and other low-grade cervical abnormalities. DNA and RNA swabs were available from these women that were no longer in the trial.

3.3.1. RNA From Cervical Swabs Total RNA from cervical samples can be extracted by employing the Qiagen RNeasy kit for total RNA following the manufacturer’s instructions. Total RNA must then be transcribed into complementary DNA (cDNA) using standard protocols before PCR amplification (20). During RNA extraction, no genomic DNA carry-over must take place. The way this is controlled is by including a DNA digestion step that removes all the genomic DNA. The total

68

Seth et al.

Fig. 2. An example of a standard calibration graph obtained for human papillomavirus (HPV) 16 E6 gene assay.

RNA is then divided into two parts—one for a reverse transcription step with reverse transcriptase enzyme, and one without any enzyme so that it acts as an internal control for the reverse transcription step. Both the cDNAs are then amplified for a housekeeping gene such as GAPDH. The cDNA without any enzyme present (known as the RT-negative control) should not show any amplification at all.

3.4. Analysis of Clinical Samples The following procedure can be carried out when analyzing the clinical samples: 1. Construct standard curves using HPV16 DNA in a 10-fold serial dilution to cover the range from 2 ng/tube to 0.0002 ng/tube as before, using a 96-well format. 2. Apply standards in duplicate in all the assays. 3. In each assay, also include negative controls (no template) to monitor any contamination. 4. Use 2 µL of cDNA from clinical samples as template in a 25-µL final volume. 5. Use individual thermal profiles for HPV 16 E2, E4, and E6, followed by dissociation curve analysis after 35–40 cycles of amplification.

On the Mx4000, the fluorescence data collected are fed into the software, which constructs a standard curve on the basis of the Ct value of each well (see Fig. 2) and a dissociation curve (melt curve) for each PCR product (see Fig. 3). Quantity of the PCR product for each of the PCR reactions is then calculated. A text report is obtained, which tabulates the quantity of the PCR product in


69

Fig. 3. Dissociation curve analysis for human papillomavirus (HPV) 16 E2, E4, and E6 gene assays, showing the three specific products amplified during the polymerase chain reaction.

ng/tube. Gene expression is expressed as a ratio of the amount of product of that particular gene divided by the amount of product for a housekeeping gene (see Note 9). The most prevalent high-risk infections are HPV16, so identification of the early gene expression will add valuable information to the clinical state of the patients with abnormal smears. The melting-curve profiles are clearly separate and distinct from each PCR, indicating that early genes can be identified and distinguished from each other using these real-time PCR techniques, avoiding any further laboratory analysis. 4. Notes 1. We order our primers from Eurogentec. The primers should be of high quality and HPLC purified. 2. QuantiTect SYBR Green PCR master mix (Qiagen UK) contains all the components for successful real-time PCRs and requires only that the concentration of the primers and the thermal cycling conditions be optimized. It contains optimized amounts of Taq DNA polymerase, a reference dye called ROX, a reporter dye (SYBR Green I), nucleotides (dNTPs), and buffer. Hotstar DNA polymerase is included in the kit; this prevents the production of nonspecific products at room temperature.

70

Seth et al.

3. SYBR Green I detects all double-stranded DNA, and it is critical to determine whether there are any primer-dimer or nonspecific products being generated in the PCR. These appear as small peaks around low temperature (approx 60–70°C). Using dissociation curve analysis for these sets of primers, no primer-dimers were detected. 4. From the information on Ct values (see Table 2) the E4 and E6gene PCRs done with 100-pmol primer dilutions produce DNA amplification between 13.52 and 15.26 cycles, whereas the E2 assay was less sensitive and had a range from 15.86 to 19.51 cycles. In the E2, E4, and E6 PCRs, the 100 pmol primer dilutions also produced small-molecular-weight byproducts (primer-dimer). 5. If the annealing temperature is too low, then the primers will bind in a nonspecific manner to other HPV types or other non-HPV sequences. Alternatively, if the annealing temperature is too high, the primers will not be able to anneal. Temperatures to be tested are 5 degrees above and below the melting temperatures (that is, the average Tm of the primers). The other components of the thermal profiles can be identical to Fig. 1. 6. Known HPV DNA types are used at a concentration of 100 pg/µL. The HPV genotype can be confirmed by sequencing. 7. Values expected are 20% for inter-assay and 10% for intra-assay. In our assays, the inter-assay variation was found to be 15%, whereas the intra-assay variation was less than 9%. 8. Analytical efficiency and specificity are determined by analysis of the standard curves and the melting-curve profiles (see Fig. 2 for an example of a standard curve and Fig. 3 for the melting-curve analysis). The gradient of the HPV 16 standard curve was between –3.330 and –3.43, which translates to an equivalent of a PCR reaction of 95–99.7% efficiency. 9. In our laboratory, GAPDH PCR product (obtained by carrying out quantitative real-time PCRs for GAPDH) is used as a housekeeping gene to normalize the results obtained for the early genes E2, E4, and E6. Another housekeeping gene that is measured regularly is betaglobin.

Acknowledgments The authors would like to acknowledge the women who took part in the TOMBOLA Trial (funded by MRC and NHS in England and the NHS in Scotland), the grant holders, and the trial staff in Grampian, Tayside, and Nottingham for the use of clinical specimens collected in this study. We also acknowledge Miss Anne Kane (Department of Histopathology, QMC, Nottingham) for her help with the diagrams. References 1. Walboomers, J. M., Jacobs, M. V., Manos, M. M., et al. (1999) Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J. Pathol. 189, 12–19.


71

2. Josefsson, A. M., Magnusson, P. E., Ylitalo, N., et al. (2000) Viral load of human papillomavirus 16 as a determinant for development of cervical carcinoma in situ: a nested case-control study. Lancet 355, 2189–2193. 3. Swan, D. C., Tucker, R, A., Tortolero-Luna, G., et al. (1999) Human papillomavirus (HPV) DNA copy number is dependent on grade of cervical disease and HPV type. J. Clin. Microbiol. 37, 1030–1034. 4. Ylitalo, N., Sorensen, P., Josefsson, A.M., Magnusson, P., Anderson, P., Ponten, J., Adami, H. O., Gyllensten, U. B., and Melbye, M. (2000) Consistent high viral load of human papillomavirus 16 and risk of cervical carcinoma in situ: a nested case-control study. Lancet 355, 2194–2198. 5. Sasieni, P., Cuzick, J., and Farmery, E. (1995) Accelerated decline in cervical cancer mortality in England and Wales. Lancet 346, 1566–1567. 6. Cuzick, J. (1998) HPV testing in cervical screening. Sexually Transmitted Infections 74, 300–301. 7. Jenkins D. (2001) Human papillomaviruses in cervical screening. Current Diagnostic Pathology 7, 96–112. 8. Manos, M. M., Ting, Y., Wright, D. K., Lewis, A. J., Broker, T., and Wolinski, S. M. (1989) Use of polymerase chain reaction amplification for the detection of genital human papillomaviruses. Cancer Cells 7, 209–214. 9. Gravitt, P. E., Peyton, C. L., Apple, R. J., and Wheeler, C. M. (1998) Genotyping of 27 human papillomavirus types by using L1 consensus PCR products by a single-hybridization, reverse line blot detection method. J. Clin. Microbiol. 36, 3020–3027. 10. Jacobs, M., Sniders, P. J. F., van den Brule, A. J. C., Helmerhorst, T., Meijers, C. J. M., and Walboomers, J. M. M. (1997) A general primer GP5+/GP6+ mediated PCR-enzyme immunoassay method for rapid detection of 14 high-risk and 6 lowrisk human papillomavirus genotypes in cervical scrapings. J. Clin. Microbiol. 35, 791–795. 11. Evander M. and Goran, W. (1991) A general primer pair for amplification and detection of genital human papillomavirus types. J. Virol. Methods 31, 239–250. 12. Kleter, B., van Doom, L. J., ter Schegget, J., et al. (1998) A novel short-fragment PCR assay for highly sensitive broad-spectrum detection of ano-genital papillomaviruses. Am. J. Pathol. 153, 1731–1739. 13. Gravitt, P. E., Peyton, C. L., Alessi, Q., et al. (2000) Improved amplification of genital human papillomaviruses. J. Clin. Microbiol. 38, 357–361. 14. Swan, D. C., Tucker, R. A., Holloway, B. P., and Icenogle, J. P. (1997) A sensitive, type-specific fluorogenic probe assay for detection of human papillomavirus DNA. J. Clin. Microbiol. 34, 886–891. 15. Whitcombe, D., Theaker, J., Guy, S. P., Brown, T.,and Little, S. (1999) Detection of PCR products using self-probing amplicons and fluorescence. Nat. Biotech. 17, 804–807. 16. Nazarenko, I. A., Bhatnagar, S. K., and Hohman, R. J. (1997) A closed tube format for amplification and detection of DNA based on energy transfer. Nucleic Acids Res. 25, 2516–2521.

72

Seth et al.

17. Jordens, J., Lanham, S., Pickett, M. A., Amarasekara, S., Abeywickrema, I., and Watt, P. J. (2000) Amplification with molecular beacon primers and reverse line blotting for the detection and typing of human papillomaviruses. J. Virol. Methods 89, 29–37. 18. Cubie, H. A., Seagar, A. L., McGoogan, E., et al. (2001) Rapid real-time PCR to distinguish between high-risk human papillomavirus types 16 and 18. J. Clin. Pathol. 54, 24–29. 19. Seth, R., Nolan, T., Rippin J, and Jenkins, D. (2004) Simultaneous detection and genotyping of HPV by quantitative real-time PCR and Sybr Green. In: A–Z of Quantitative PCR, Bustin, S. (ed.) La Jolla, CA: International University Line. 20. McLaughlan, J., Seth, R., Vautier, G., et al. (1997). Interleukin-8 and inducible nitric oxide synthase mRNA levels in inflammatory bowel disease at first presentation. J. Pathol. 181, 87–92.

p16INK4a and Integrated Genomes as Progression Markers

73

7 Analysis of p16INK4a and Integrated HPV Genomes as Progression Markers Svetlana Vinokurova, Nicolas Wentzensen, and Magnus von Knebel Doeberitz Summary Most cervical cancers are preventable when the precursor lesions are detected in time. Human papilloma viruses (HPVs) are the main risk factors for cervical cancer development, but there is a high percentage of healthy women infected with HPV that never develop a lesion. Only a small percentage of low-grade dysplasias finally grow out to invasive cancer. Several biomarkers can be used to identify lesions at risk for malignant progression. Overexpression of p16INK4a is induced by the viral oncoprotein E7 and distinguishes dysplastic lesions from benign changes. Integration of human papillomavirus DNA into the host genome is mainly found in high-grade dysplastic lesions and invasive cancers, and points to an increased progression potential.

1. Introduction The principal carcinogenic factors of high-risk human papillomaviruses (HR-HPV) are the viral oncogenes E6 and E7, which code for proteins interfering substantially with apoptosis and cell-cycle regulation. Expression of these viral genes is required to induce and maintain cervical carcinogenesis. Most of the acute HR-HPV infections regress spontaneously within a couple of months (1). However, in a few cases (3 to 10%), such acute HR-HPV infections might persist for a longer period of time. This induces incompatible biochemical signals that regulate replication of the host cell and the viral genome, and rapidly leads to chromosomal instability (2). The most important interactions of E6 and E7 are with p53 and pRB, both cellular tumor-suppressor genes (3). E6 binds to p53 and leads to its degradation. When p53 function is lost, chromosomal damage accumulates in the cells and can lead to malignant transformation. pRB is an inhibitor of G1-S transition. When E7 is bound From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

73

74

Vinokurova et al.

Fig. 1. Mechanism of p16INK4a overexpression. (A) E2F is bound by Rb, no expression of cell-cycle promoting genes. (B) Phosphorylation of cyclin-dependent kinases (CDK) leads to the release of E2F and to the activation of cell-cycle promoting genes. p16INK4 down-regulates CDK action to maintain strict cell-cycle regulation. (C) HR-HPV E7 is binding to Rb independent of the phosphorylation state; activation of p16INK4a has no effect on the E2F release.

to pRB, E2F is released and leads to the activation of several cell-cycle promoting factors (4). Another independent effect of E6 and E7 expression is the major disturbance of the mitotic spindle apparatus, leading to abnormal mitoses with multipolar mitotic figures, which results in severe numeric and chromosomal aberrations (2). The interaction of E7 with the pRB gene product results in release of E2F from the active pRB-E2F complex and premature degradation of the pRB-E7 complex (4). Because pRB-E2F inhibits transcription of the p16INK4a gene, expression of HR-HPV E7 results in excessive and deregulated transcription and translation of the p16INK4a (p16) gene (5) (Fig. 1). Importantly, this overexpression of p16INK4a due to the expression of the HR-HPV E7 gene product is not observed when the E7 protein is normally expressed at low levels during an acute HR-HPV infection (6). Because p16INK4a is not expressed in normal cervical squamous epithelia, screening for p16INK4a overexpressing cells allows specific identification of dysplastic lesions in histological sections and in cervical smears (7–11).


75

Only a minority of patients that have developed a cervical dysplasia have lesions that finally progress to invasive cancer. Identification of molecular markers that reliably predict the progression risk of cervical lesions might contribute to disease management of cervical dysplasia. It was observed that a fraction of high-grade dysplastic lesions as well as the majority of cervical cancers harbor HPV DNA integrated into the host genome (15). It was shown that integration of HR-HPV genomes in cervical lesions results in enhanced expression of the viral oncogenes E6 and E7, and thus confers a strong promoting factor in the progression of dysplastic cervical lesions (12,13). It was shown that chromosomal instability and aneuploidization of the host cell genome precedes integration of the viral genome, suggesting that integration reflects the severe chromosomal damage that occurs during the progression of the dysplastic lesion (14). HPV DNA is integrated into the host genome in over 80% of the cervical carcinomas (15). There is no specific integration hot spot—every HPV integration happens at unique sites distributed over the whole human genome (15,16). However, so-called fragile sites within the human genome appear to facilitate the integration of HR-HPV genomes. As a result of the uniqueness of every integration site, integration detection can be used as a highly specific and individual tumor marker. The precise recombination sites of viral and cellular sequences represents a molecular fingerprint that is applied to delineate the clonality of a lesion, to monitor patients after treatment for local recurrences, or to monitor distant metastases and lymph nodes using highly sensitive and specific polymerase chain reaction (PCR) methods. Integration of HR-HPV genomes in cervical carcinoma cells usually results both in disruption of the E1 or E2 open reading frames and in disruption or deletion of the viral early-region polyadenylation signal from the viral oncogene-encoding sequences. Consequently, transcripts derived from the integrated E6 and E7 oncogenes commonly encompass viral sequences at their 5'-ends and flanking cellular sequences at their 3'-ends. These structural differences of integrated or episomal viral oncogene transcripts are detectable by the amplification of papillomavirus oncogene transcripts (APOT) assay. Using an oligo(dT)17-primer coupled to a linker sequence ([dT]17-p3), the reverse transcription (RT) of all of the mRNAs is initiated by binding to their poly(A) tail. Subsequently, both episome- and integrate-derived HPV oncogene transcripts are amplified by nested PCR reactions using E7-specific forward primers in combination with primers p3 and (dT)17-p3, respectively (Fig. 2). Integrate-derived transcripts can now be differentiated from the abundant episome-derived transcript (1050 bp) because of their different sizes. Furthermore, the obtained PCR fragments can be verified by Southern blot hybridization analysis using HPV E7- and E4-specific oligonucleotides (Fig. 3, h1 and h2).

76

Vinokurova et al.

Fig. 2. APOT diagram. Arrows: Primer binding sites. P1, P2: HPV-specific forward primers for first and second PCR. dT: Oligo dT primer. P3: Adaptor primer. H1: E7 probe. H2: E4 probe.


77

The two methods described in this chapter were developed to detect different stages in the progression of HR-HPV-induced lesions. p16INK4a staining marks those persistently HR-HPV-infected cells that display deregulated expression of the viral oncogenes. This starts in early dysplastic lesions and helps to distinguish dysplastic cells in equivocal cytological or histological slides or samples from benign, inflammatory, and metaplastic lesions. However, only a fraction of these p16ink4a-positive lesions will grow out to invasive carcinomas. Detection of integrated viral genomes or their transcription, in contrast, points to more advanced lesions (CIN3 and carcinoma) with a very high potential of further malignant progression. 2. Materials 2.1. p16 Staining 1. Ethanol: 95%, 70%, 50%. 2. Xylene. 3. Phosphate-buffered saline (PBS): 137 mM NaCl, 2.7 mM KCl, 1.5 mM KH2PO4, 8.1 mM Na2HPO4 (pH 7.4) + 0.1% Tween-20 (PBS + T). 4. Slides (e.g., SuperFrost Plus, poly-L-lysine-coated slides). 5. Immunohistochemistry Marking Pen (e.g., DAKO Pen, DakoCytomation, Glostrup, Denmark). 6. Primary antibody (e.g., E6H4 by MTM Laboratories) or appropriate p16 staining kit (CINTec histology/cytology kits, Dakocytomation, Glostrup, Denmark). 7. Appropriate biotin-coupled secondary antibody (anti-mouse, e.g., Vectastain, Vector Laboratories, Burlingame, CA). 8. Appropriate AB complex and substrate amino-ethylcarbazol (AEC) or diaminobenzidine (DAB) (e.g., Vectastain kits, Vector Laboratories, Burlingame, CA). 9. Counterstain: hematoxylin, (e.g., Mayer’s hematoxylin, DakoCytomation, Glostrup, Denmark). 10. Coverslips. 11. Humid chamber. 12. Water bath with lid (for epitope retrieval at 95–99°C). 13. Light microscope. 14. Mounting medium (e.g., aqueous Aquatex, Merck, or glycerol gelatin). 15. Positive and negative specimens to use as process controls.

Fig. 3. (opposite page) Amplification and hybridization results (human papillomavirus [HPV] 16). Upper row: agarose gel electrophoresis after amplification of papillomavirus oncogene transcripts (APOT) reverse-transcription polymerase chain reaction. Middle and lower rows: Southern blot hybridization with HPV E7/E4-specific probes. 1–3 = normal; 4–6 = CIN1; 7–9 = CIN2; 10–12 = CIN3; 13–15 = carcinomas. * Integrate detection by differential hybridization.

78

Vinokurova et al.

2.2. Amplification of Papillomavirus Oncogene Transcripts 1. RNeasy (Qiagen). 2. RNAlater (Ambion, Woodward, TX). 3. Enzymes and buffers: SuperScript™ II, RNase H- Reverse Transcriptase (Invitrogen). 4. Oligonucleotide primers: (dT)17-P3: GACTCGAGTCGACATCGA TTTTTTTTTTTTTTTTT P3: GAC TCG AGT CGA CAT CG HPV16-P1: CGG ACA GAG CCC ATT ACA AT HPV18-P1: TAG AAA GCT CAG CAG ACG ACC HPV16-P2: CCT TTT GTT GCA AGT GTG ACT CTA CG HPV18-P2: ACG ACC TTC GAG CAT TCC AGC AG 5. Whatman filter paper. 6. Nylon membranes (e.g., Hybond N+, Amersham Life Science). 7. Gel Extraction Kit (Qiagen). 8. Agarose and DNA sequencing gel equipment. 9. ECL oligolabeling and detection kit (Amersham). 10. Hybridization probes H1-16: TCGTACTTTGGAAGACCTGTTAATG H1-18: GTTTCTGAACACCCTGTCCTTTGTG H2-16: GAAGAAACACAGACGACTATCCAG H2-18: CAGCTACACCTACAGGCAACAACAA 11. TA cloning kit (Invitrogen).

3. Methods 3.1 p16INK4a Histology (see Note 1)

3.1.1. Tissue Preparation Tissues fixed with neutral buffered formalin and embedded in paraffin can be used for p16 staining. 1. Cut paraffin-embedded tissues into 1–2-µm slices and place on glass slides. 2. To dry, incubate slides overnight at 37°C. 3. For proper staining, paraffin has to be removed completely by washing in xylene twice for 5 min, 95% ethanol twice for 3 min, 70% ethanol three times for 3 min, and finally distilled water once for >30 s. (see Note 2). 4. For antigen retrieval, incubate slides in Tris-ethylenediamine tetraacetic acid (EDTA) (pH 9.0) at 95°C (water bath) for 10 min. Cool down to room temperature for 20 min. Apply marking pen to confine staining solutions (see Note 3). 5. To block, incubate sections with 1% peroxidase for 5 min. 6. Wash in PBS + T for 5 min.


79

3.1.2. Staining and Mounting 1. Incubate sections with primary antibody (for E6H4 use a concentration of 0.6 µg/mL) in the appropriate blocking solution (for E6H4 use 2% horse serum in PBS + T) at room temperature for 30 min. 2. Wash in PBS + T twice for 5 min. 3. Incubate with biotinylated secondary antibody (e.g., horse-anti-mouse immunoglobulin [Ig]G, 1:200) in the appropriate blocking solution (e.g., 2% horse serum in PBS + T) at room temperature for 30 min. 4. Wash in PBS + T twice for 5 min. 5. Dilute equal amounts of avidin- and biotin-coupled horseradish peroxidase solution 1:50 in PBS + T and incubate sections at room temperature for 30 min. 6. Wash in PBS + T twice for 5 min. 7. Use either AEC or DAB substrates for color reaction according to the manufacturer’s protocol. 8. Wash in water at room temperature for 5 min. 9. Counterstain nuclei with hematoxylin. According to the strength of the hematoxylin solution, incubate slides for 1–5 min. 10. Blue sections in tap water for 5–10 min. 11. Mount sections in the appropriate mounting medium: for DAB substrates, use aqueous mounting medium, for AEC use glycerol gelatin (see Notes 4 and 5).

3.2. p16INK4a Cytology 3.2.1. Cytology Slide Preparation Different thin-layer cytology systems can be used for the procedure, e.g., Cytyc, SEROA, Autocyte, and so on. The slides should be ethanol (100%) fixed after preparation. Follow the steps outlined under Subheading 3.1.1., beginning with antigen retrieval.

3.3. Amplification of Papillomavirus Oncogene Transcripts 3.3.1. RNA Isolation and Quality RNA integrity is very important for good performance of the APOT assay. When RNA integrity is assured, APOT can be performed from very small amounts of clinical material, such as cervical swabs or small biopsies. It is very important to stabilize RNA immediately after sample extraction; for optimal results, samples should be frozen in liquid nitrogen (see Note 6). All RNA isolation methods show sufficient results with APOT amplification when good RNA was used as starting material (see Note 7).

80

Vinokurova et al.

3.3.2. DNAse Digest If problems with DNA contamination occur, a DNAse digest can be performed prior to reverse transcription. However, a general application of DNAse does not seem to be necessary (see Note 8).

3.3.3. Reverse Transcription Use total RNA (1 ng to 1 µg) for reverse transcription with an oligo(dT) 17-primer coupled to a linker sequence (dT)17-p3 (see Note 9). 1. Prepare a master mix (1) containing per reaction: 4 µL (1 ng to 1 µg) of total RNA template, 7 µL water, 1 µL 25 µM (dT)17-p3. 2. Prepare a master mix (2) containing per reaction: 4 µL 5X RT buffer, 2 µL 0.1 M dithiothreitol (DTT), 1 µL 10 mM dNTP, 0.1–0.2 µL (20–40 U) MMLV RT SuperScript. 3. Heat-denature master mix 1 at 70°C for 10 min and quickly chill on ice. 4. Add master mix 2 and incubate at 42°C for 60 min. 5. Inactivate for 5 min at 90°C.

3.3.4. PCR Amplification 1. Prepare a first PCR reaction containing the following: 1X PCR buffer; 0.2 mM dNTPs; 1.5 mM MgCl2; 0.25 µM primer HPV16-P1 (for HPV16) or primer HPV18-P1 (for HPV18), and 0.25 µM P3; 5 U Taq polymerase; and 4 µL template; in a total of 50 µL. 2. Cycle according to the following conditions: Initial denaturation for 3 min at 94°C, then 30 cycles of denaturation for 40 s at 94°C; annealing for 30 s at 59°C (for HPV16) or 61°C (HPV18); extension for 4 min at 72°C. The reactions finish with a final extension of 7 min at 72°C. 3. Prepare a second PCR reaction containing the following: 1X PCR buffer; 0.2 mM dNTPs; 1.5 mM MgCl2; 0.25 µM primer HPV16-P2 (for HPV16) or primer HPV18-P2 (for HPV18), and 0.25 µM primer d(T)17-P3; 5 U Taq polymerase; and 4 µL of the first PCR product; in a total volume of 50 µL. 4. Cycle according to the following conditions: Initial denaturation for 3 min at 94°C, then 30 cycles of denaturation for 40 s at 94°C; annealing for 30 s at 67°C (for HPV16) or 70°C (for HPV18); extension for 4 min at 72°C. The reactions finish with a final extension of 7 min at 72°C.

3.3.5. Hybridization 1. Electrophorese the PCR products in 1.2% agarose gels and transfer onto nylon membranes. 2. Hybridize with an E7-specific probe (H1) at 50°C. Hybridize a second parallel filter with an E4-specific probe (H2) at 50°C to highlight amplimeres that encompass E4 sequences.


81

3. Use ECL oligolabeling and detection kit for labeling and detection of the probes according to the manufacturer’s instructions. Alternatively, other ECL detection systems can be used according to your personal preferences.

3.3.6. Interpretation of Results All specific amplimers show hybridization signals with the E7 probe. Amplimeres that do not hybridize with the E4-specific probe or that display a different size than the major E7-E1^E4 episomal transcript (approx 1050 bp in length for HPV16 and 1000 bp for HPV18) are suspected to be derived from integrated HPV genomes (see Note 10).

3.3.7. Sequencing Excise PCR products of interest from the gel and extract using Gel Extraction Kit. PCR products can be cloned, for example into pCR2.1 vector using the TA cloning Kit (Invitrogen), or directly sequenced using P2 primers. 4. Notes 1. The authors are experienced users of the p16INK4a mouse monoclonal antibody E6H4 (MTM laboratories, Heidelberg). DAKO is offering staining kits for p16INK4a histology and cytology using modified and shortened protocols compared to the general staining protocols described in this chapter. Other antibodies may be used, but staining procedures have to be adjusted accordingly. 2. Change solutions after 30–40 slides. 3. Touching tissue and cells with the marking pen might result in staining artifacts. Leave enough space between tissue/cells and marking pen. 4. A good nuclear counterstain is necessary for the assessment of nuclei to discern rare p16-positive metaplastic cells from dysplastic cells showing nuclear atypias. 5. It is of utmost importance that the slides do not dry out during the staining procedure. As with all immunohistochemistry procedures, staining in the border areas of slides is not considered to be specific. In every staining procedure, positive controls should be included, either confirmed p16-positive tissue samples or p16expressing cell lines (e.g., HeLa cells). To avoid contamination of cytological slides with detached cells from the positive-control slide, these samples should be washed in different containers. 6. RNA stabilization solutions can be used as a substitute for liquid nitrogen. Good APOT amplification results can be obtained when samples are immediately transferred to RNAlater (Ambion) and stored up to 1 wk at 25°C and up to 1 mo at 4°C. Long-term storage is possible in RNAlater solution at –20°C or –80°C. The quality of the isolated RNA should be determined by amplification of housekeeping gene mRNAs like GAPDH or beta-actin. 7. Various RNA preparation methods can be used, including modified phenol/chloroform assays, Trizol protocols, or column-based methods, like RNeasy (Qiagen).

82

Vinokurova et al.

8. If RNA is extracted using the RNeasy kit, DNA can also be isolated from the first wash flow-throughs. The isolated DNA can be used for HPV typing or genomebased integration detection. DNAse digestion of the isolated RNA is generally not performed. 9. SuperScript II RNase H- Reverse Transcriptase (Invitrogen) can be used to synthesize first-strand cDNA and will generally give higher yields of cDNA and more full-length product than other reverse transcriptases. 10. When high amounts of episomal transcripts are present, mispriming of the oligo(dT) primer to A-rich regions can lead to amplification episomal fragment that have a different length from the standard episomal transcript (1050 bp for HPV16 and 1000 bp for HPV18).

References 1. Ho, G. Y., Bierman, R., Beardsley, L., Chang, C. J., and Burk, R. D. (1998) Natural history of cervicovaginal papillomavirus infection in young women. N. Engl. J. Med. 338, 423–428. 2. Duensing, S. and Munger, K. (2004) Mechanisms of genomic instability in human cancer: insights from studies with human papillomavirus oncoproteins. Int. J. Cancer 109, 157–162. 3. Mantovani, F. and Banks, L. (2001) The human papillomavirus E6 protein and its contribution to malignant progression. Oncogene 20, 7874–7887. 4. Munger, K., Basile, J. R., Duensing, S., et al. (2001) Biological activities and molecular targets of the human papillomavirus E7 oncoprotein. Oncogene 20, 7888–7898. 5. Khleif, S. N., DeGregori, J., Yee, C. L., et al. (1996) Inhibition of cyclin D-CDK4/ CDK6 activity is associated with an E2F-mediated induction of cyclin kinase inhibitor activity. Proc. Natl. Acad. Sci. USA 93, 4350–4354. 6. Von Knebel Doeberitz, M. (2002) New markers for cervical dysplasia to visualise the genomic chaos created by aberrant oncogenic papillomavirus infections. Eur. J. Cancer 38, 2229–2242. 7. Klaes, R., Woerner, S. M., Ridder, R., et al. (1999) Detection of high-risk cervical intraepithelial neoplasia and cervical cancer by amplification of transcripts derived from integrated papillomavirus oncogenes. Cancer Res. 59, 6132–6136. 8. Klaes, R., Friedrich, T., Spitkovsky, D., et al. (2001) Overexpression of p16(INK4A) as a specific marker for dysplastic and neoplastic epithelial cells of the cervix uteri. Int. J. Cancer 92, 276–284. 9. Sahebali, S., Depuydt, C. E., Segers, K., et al. (2004) p16INK4a as an adjunct marker in liquid-based cervical cytology. Int. J. Cancer 108, 871–876. 10. Trunk, M. J., Dallenbach-Hellweg, G.., Ridder, R., et al. Morphological characteristics of p16ink4a positive cells in cervical cytology samples. Acta Cytologica 48, 771–782. 11. Klaes, R., Benner, A., Friedrich, T., et al. (2002) p16INK4a immunohistochemistry improves interobserver agreement in the diagnosis of cervical intraepithelial neoplasia. Am. J. Surg. Pathol. 26, 1389–1399. 12. Jeon, S., Allen-Hoffmann, B. L., and Lambert, P. F. (1995) Integration of human


13.

14.

15.

16.

83

papillomavirus type 16 into the human genome correlates with a selective growth advantage of cells. J. Virol. 69, 2989–2997. Jeon, S. and Lambert, P. F. (1995) Integration of human papillomavirus type 16 DNA into the human genome leads to increased stability of E6 and E7 mRNAs: implications for cervical carcinogenesis. Proc. Natl. Acad. Sci. USA 92, 1654–1658. Melsheimer, P., Vinokurova, S., Wentzensen, N., Bastert, G., and von Knebel Doeberitz, M. (2004) DNA aneuploidy and integration of HPV-16 E6/E7 oncogenes in intraepithelial neoplasia and invasive squamous cell carcinoma of the cervix uteri. Clin. Cancer Res. 10,3059–3063. Wentzensen, N., Vinokurova, S., and von Knebel Doeberitz, M. (2004) Systematic review of genomic integration sites of human papillomavirus genomes in epithelial dysplasia and invasive cancer of the female lower genital tract. Cancer Res. 64, 3878–3884. Ziegert, C., Wentzensen, N., Vinokurova, S., et al. (2003) A comprehensive analysis of HPV integration loci in anogenital lesions combining transcript and genomebased amplification techniques. Oncogene 22, 3977–3984.

Use of Biomarkers in the Evaluation of CIN

85

8 Use of Biomarkers in the Evaluation of CIN Grade and Progression of Early CIN Jan P. A. Baak and Arnold-Jan Kruse Summary The treatment of cervical intraepithelial neoplasia (CIN) depends on the evaluation of CIN grade. The grading of CIN is however problematic, as intra- and interobserver reproducibility of CIN-grade evaluation among pathologists is not perfect. There are also difficulties in reliably distinguishing CIN from nonneoplastic lesions, and over- or undertreatment can be the result. These points suggest a need for adjuvant methods that can distinguish CIN from nonneoplastic lesions, and can distinguish different CIN grades and predict the risk of progression of early CIN1 and -2 lesions. This chapter describes the use of biomarker-related methods for the diagnosis and prognostic evaluation of patients with CIN1 and CIN2. As CIN involves the progressive dysfunction of proliferation and differentiation activities in cervical epithelial cells, we have concentrated in this chapter on demonstrating the utility of proliferation- and differentiation-related biomarkers.

1. Introduction Treatment of cervical intraepithelial neoplasia (CIN) depends on the accurate assessment of CIN grade. However, CIN grade assessment is problematic (1), as intra- and interobserver reproducibility among pathologists is not perfect (2–7). There are also difficulties in reliably distinguishing CIN from nonneoplastic lesions (6–9), and over- or undertreatment can be the result. These points indicate a need in surgical pathology and gynecology practices for adjuvant methods that can differentiate between CIN grades and distinguish CIN from non-neoplastic lesions. Such methods should also predict the risk of finding high-grade CIN3, whether in the same diagnostic biopsy, in the cervical mucosa left in situ adjacent to the biopsy, or in the follow-up of early CIN1 and -2 lesions (“early CIN”). This chapter describes the application of adjuvant methods developed for this purpose. For too long, such methods have been mostly morphological in nature and have not, in any adequate way, incorpoFrom: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

85

86

Baak and Kruse

rated knowledge of the underlying biological mechanisms leading to or explaining the disease. As CIN involves progressive dysfunction of the proliferation and differentiation activity of cervical epithelial cells, we have concentrated on studying proliferation- and differentiation-related biomarkers. There is considerable variation in the literature regarding the value of biomarkers, although certain trends are now generally accepted. The reasons for these discrepancies are manifold. First, most biochemical sampling methods carry the risk of mixing lesions with normal tissue. Depending on how much of each is mixed, the results can vary considerably. Immunohistochemical (IHC) methods can avoid this disadvantage, but the results may vary as a result of staining differences and interpretation differences. We use a strict staining procedure to limit such variations (10). To avoid interpretational errors, quantitative objective methods can be used in order to improve the reproducibility of assessments. In addition to the objective assessment of certain features, the detection of differences that escape subjective judgment is also possible with quantitative microscopy (11). Finally, assessment of the different features can be performed in the deep, middle, and superficial layers of the epithelium. As images of the cervical epithelium are a snapshot of a highly dynamic process, this can be done as follows. Immature cells (large nuclei, little cytoplasm) are “born” in the deep epithelium and, while moving to the surface, mature into highly differentiated cells (small pyknotic nuclei, much cytoplasm). This process may take 7–14 d, at the end of which the cells are desquamated. Biomarkers reflect these maturation processes, and it can therefore be expected that they will vary greatly in the different layers of the epithelium. Figure 1A illustrates that for the Ki67 nuclear proliferation-associated antigen, this is indeed the case. Consequently, their analysis in the epithelium as a whole is like mixing apples and oranges. An average epithelial value therefore is of little significance, and geographyspecific interpretation is essential to catch the biology of the CIN lesion. In order to obtain an adequate impression of the biomarker dynamics in normal and neoplastic epithelium, the biomarkers must be quantified separately in the different layers of the epithelium (see Fig. 1B). This chapter gives a summarized description of each method. 2. Materials 2.1. Staining Procedures 1. 2. 3. 4. 5.

Silanized slides (Dako, Glostrup, Denmark). Xylene. Graded series of alcohol solutions. 10 mM Tris-HCl, 1 mM ethylenediamine tetraacetic acid (EDTA) (pH 9.0). Autostainer (DAKO, Glostrup, Denmark).


87

Fig. 1. (A) Example of a human papillomavirus (HPV)-positive but morphologically normal epithelium. Left: hematoxylin and eosin (H&E); inset: normal basal and parabasal nuclei. Right: Ki67 (note the parabasal location and low number of positive nuclei, arrows). (B) Diagram illustrating the different epithelial layers in which the measurements were performed. 6. 7. 8. 9.

TBS (S1968), 0.05%. Tween-20 (pH 7.6). Peroxidase blocking reagent S2001 (DAKO, Glostrup, Denmark). Monoclonal antibodies. Table 1 gives an overview of the clones, dilutions, and manufacturers of each of the monoclonal antibodies used. 10. DAKO antibody diluent S0809.

88

Baak and Kruse

Table 1 Clones, Dilutions, and Manufacturers of the Monoclonal Antibodies Used Monoclonal antibody CK-13 CK-14 cyclin D1 cyclin E hTERT p16 p21 pRb cyclin A p27 p53 Ki67

Clone KS-1A3 LL002 P2D11F11 13A3 44F12 6H12 4D10 13A10 CYA06 SX53G8 DO-7 MIB-1

Dilution 1:200 1:40 1:25 1:40 1:100 1:25 1:25 1:25 1:40 1:100 1:200 1:100

Manufacturer Novocastra, Newcastle upon Tyne, UK Novocastra Novocastra Novocastra Novocastra Novocastra Novocastra Novocastra Neomarkers, Fremont, USA DAKO, Glostrup, Denmark DAKO DAKO

11. Peroxidase/diaminobenzidine (DAB) (K 5007) (ChemMate Envision Kit, DAKO, Glostrup, Denmark. 12. Envision/HRP, rabbit/mouse (Envision = ENV). 13. DAB + chromogen. 14. Hematoxylin.

2.2. Immunoquantitation and Syntactic Structure Analysis Interactive image analysis system (QPRODIT). 3. Methods 3.1. Staining Procedures 1. Mount paraffin sections of 4-µm thickness adjacent to the hematoxylin and eosin (H&E) sections used for the CIN grade assessment, onto silanized slides. 2. Dry overnight at 37°C followed by 1 h at 60°C. 3. Deparaffinize the sections in xylene and rehydrate in a graded series of alcohol solutions. 4. Perform antigen retrieval by pressure cooking in 10 mM Tris, 1 mM EDTA (pH 9.0) for 2 min at full pressure followed by cooling for 15 min. 5. Use an autostainer for the immunostaining. 6. Add TBS (S1968) at 0.05% and Tween-20 (pH 7.6) as the rinse buffer. 7. Block endogenous peroxidase activity by incubating for 10 min in peroxidaseblocking agent. 8. Incubate with monoclonal antibodies at the dilutions shown in Table 1. 9. Visualize the immune complex by peroxidase/DAB with incubation of Envision/ HRP, rabbit mouse (ENV) for 30 min, and DAB + chromogen for 10 min. 10. Counterstain the section with hematoxylin.


89

11. To ensure the presence of the same CIN lesion in all test sections, cut the section adjacent to the sections used for immunostaining and stain with H&E.

3.2. Ki67 Immunoquantitation Using MIB-1 Antibody 1. Mark in each case the area with the subjectively highest CIN grade. 2. Avoid technically unsatisfactorily cases for quantitative Ki67 analysis (see Note 1). 3. Demarcate with the cursor of the interactive image analysis system, the lumen and basal membrane of the most severely dysplastic epithelium (making sure that the length of any one epithelial strip is at least 30 µm). 4. At a final monitor magnification of ×1400, mark the center of each Ki67-positive nucleus using the cursor of the system. Figure 2 illustrates this method. 5. The system automatically calculates multiple quantitative features (including various descriptive statistics for each feature) in each case. These include the distance between the nucleus and basement membrane (=DBM), epithelial thickness (=T) at the location of the nucleus indicated, distance between the nucleus and the lumen (=DL), stratification index (Si) (=DBM/T, which is the distance between the nucleus and basal membrane, divided by the epithelial thickness), total number of Ki67-positive nuclei per 100 µm basement membrane, and percentage Ki67-positive nuclei in the deep third, the middle third, and the upper third of the epithelium. 6. The QPRODIT system also calculates a large number of descriptive statistics for each quantitative variable, such as mean, median, 5th, 10th, . . . 90th, 95th percentiles.

3.3. Evaluation of MIB-1-Positive Cell Cluster Criterion 1. Assess the presence of a MIB-1 (Ki67) positive cell cluster in cervical squamous epithelium. The MIB-1 positive cell cluster criterion is defined as positive when a cluster of at least two strongly stained adjacent epithelial nuclei are present in the upper two-thirds of the epithelial thickness anywhere within the lesion (12). Figure 1A shows a human papillomavirus (HPV)-positive but morphologically normal epithelium. Figure 3 shows an HPV-positive MIB-1-cell-cluster positive CIN lesion. It is important to exclude inflammatory cells and tangentially cut parabasal epithelium (see Note 2).

3.4. Evaluation of Immunopositivity 1. Regard cells as either negative or positive and assess the percentage of positive cells in the basal cell layer, the lower half (excluding the basal cells), and the upper half of the epithelium by two independent observers. 2. In case of interobserver disagreement, obtain consensus through discussion.

3.5. Syntactic Structure Analysis 1. Select the most severely dysplastic part of the epithelium in the H&E section. 2. Mark in this area the upper border, the basal membrane, and the borders between the two deepest layers of the epithelium (the deepest layer excluding the basal cell layer).

90

Baak and Kruse

Fig. 2. Illustration of the Windows-based image analysis method. The microscopic image of the epithelium is shown on the monitor of the image-analysis system. Using the mouse the operator demarcates a diagnostic epithelium strip, carefully avoiding tangentially cut areas. The demarcation lines are shown as white lines (surface, basal membrane, left, right). The operator then clicks the mouse on all Ki67-positive nuclei within the demarcated strip. After each click, the system automatically draws a perpendicular line from that point to the basal membrane and over the full thickness of the epithelium (these thin dotted lines are barely visible for all nuclei) and calculates several quantitative features, such as thickness (=T) of the epithelium at that point, distance (=D) of the point to the basal membrane and stratification index (SI = D/T). These quantitative features are shown in the left panel after each click and are stored automatically. In the figure, lines D (dotted lines) and T (continuous lines) are emphasized for three nuclei at A-A', B-B', and C-C'. The SIs are 0.90, 0.45, and 0.19 respectively. The image analysis program automatically calculates per sample many quantitative features, of which SI90 and MIDTHIRD are the most important. The SI90 is the 90th percentile of the stratification index. The MIDTHIRD is the percentage of all MIB-1-positive nuclei in the whole thickness of the middle third of the epithelium. 3. Mark on a video screen, by setting a point with the cursor of the interactive image analysis system, the centers of gravity of all nuclei in the lower deep half of the epithelium in five fields of vision chosen randomly. Figure 4 illustrates this image-analysis method.


91

Fig. 3. Example of a cervical intraepithelial neoplasia (CIN) 1 lesion (left: hematoxylin and eosin [H&E]) with (right) MIB-1-positive cell clusters (arrows).

The computer program composes the Voronoi diagram (VD) and the minimum spanning tree (MST) using this set of points. The Voronoi tessellation splits the image plane into polygons, each containing one nuclear center of gravity in its center, and in such a way that each point in the polygon is closer to this center nucleus than to any other point in the plane. The MST interconnects a set of nuclei in such a way that the total length of the lines was minimal and no loops were formed. Figure 4 illustrates this image-analysis method.

3.6. Use of Biomarkers in Daily Routine On the basis of the above-mentioned methods, we recommend the following strategy for the handling of a cervical biopsy in a surgical pathology laboratory, in the realm of CIN analysis: 1. Analyze the diagnostic H&E-stained section for routine evaluation. 2. Scan the serial section after staining for p16 (see Chapter 7) in order to identify diffusely positive squamous areas. Based on recent studies in the literature and our own routine use of p16 (unpublished results), these are nearly always dysplastic (false-positive staining for p16 is very rare and easily recognized) (11). The underlying cause of diffuse p16 positivity is usually hrHPV positivity of the p16-positive squamous cells.

92

Baak and Kruse

Fig. 4. Illustration of the syntactic structure analysis (SSA). In each case, the most severely dysplastic part of the epithelium in the hematoxylin and eosin (H&E) section used for the routinely assessed diagnosis was selected. In this area, the upper border, the basal membrane, and the borders between the two deepest layers of the epithelium were marked (the deepest layer excluding the basal cell layer). SSA was performed in fields of vision chosen randomly in the already demarcated area by constructing a Voronoi diagram (VD) and a minimum spanning tree (MST) per field. Briefly, with a ×40 objective (final magnification ×1200) the centers of gravity of all nuclei in the lower deep half of the epithelium per field of vision were interactively marked on a video screen by setting a point with the cursor of the system. Using this set of points, the computer program composed the VD and the MST. The Voronoi tessellation splits the image plane into polygons, each containing one nuclear center of gravity in its center and in such a way that each point in the polygon was closer to this center nucleus than to any other point in the plane. The MST interconnected a set of nuclei in such a way that the total length of the lines was minimal and no loops were formed. The system automatically calculates multiple quantitative features (including various descriptive statistics for each feature) in each case derived from the VD. These include (1) total points clicked on with the cursor, (2) the sum of all polygon surfaces, (3) average surface of the polygons, (4) minimum surface of the polygons, (5) maximum surface of the polygons, (6) standard deviation of the polygons, (7) average of all roundness factors, (8) standard deviation of all roundness factors, (9) area disorder, (10) the number and percentage of selected points from which the surrounding surface


93

3. Evaluate further with Ki67 in the next serial section. Ki67-positive cell clusters further indicate CIN (12,13). 4. Perform quantitative Ki67 analysis for objective grading support (Fig. 5A) and progression risk indication (Fig. 5B,C), in the case of CIN1 and CIN2 (see Fig. 2 for a detailed description of the method and the most important features used). If Ki67 Si90 exceeds 0.57 and/or MIDTHIRD exceeds 30%, the likelihood of CIN3 in the follow-up is high (30%) (see Note 3 and Fig. 5B,C). Figure 5B,C shows the Ki67-only models used at our laboratory to grade and predict the progression risk of CIN lesions (14–18). 5. Then, in the subsequent section, which is stained for retinoblastoma protein (Rb), analyze the Rb positivity of nuclei in the lower half of the epithelium. The retinoblastoma (Rb) gene was the first suppressor gene discovered. pRb, the protein product of Rb, is a nuclear phosphoprotein that plays a key role in regulating the cell cycle. In its active state, pRb serves as a brake on the advancement of cells from the G1 to the S phase of the cell cycle. It is thus understandable that a decrease of pRb in the lower cell layers of the cervical epithelium will lead to an increased (uncontrolled) proliferation. This then may be reflected as an increase in Ki-67-positive nuclei in the middle and upper cell layers. If the combination of Ki67-Si90 > 0.57 occurs together with Rb < 40% (which means that in the lower half of the epithelium the percentage of pRb-positive nuclei is less than 40%), progression risk in CIN1 and CIN2 is very high (approx 50%). Figure 5D illustrates this graphically (19). 6. Finally, analyze CK13 and CK14. These features have prognostic value, but only in the high-risk subgroup of Ki67 Si90 > 0.57 with Rb < 40%. Combined CK13 < 80% and CK14 15 min. 4. Centrifuge the samples at at least 10,000g for 15 min, 4°C. Remove ethanol, taking care not to dislodge the pellets. Pellet again and carefully remove the last traces of ethanol. Dry 5 min at room temperature. 5. Resuspend the pellets in 10 µL of hybridization buffer. Vortex each for 5–10 s, then centrifuge briefly to collect droplets to the bottom of the tube. 6. Incubate at approx 90°C (±5°C) for 3–4 min to denature RNA and aid in solubilization, then revortex and pellet briefly. 7. Incubate tubes for 2–18 h, preferably in a 42–45°C cabinet-type incubator, or in a 45°C water bath or heat block (see Note 12).

3.3. RNase Digestion of Hybridized Probe and Sample RNA 1. Thaw RNase digestion buffer and make RNase solution (150 µL × number of samples) at 1:100 dilution of RNase. 2. Spin samples briefly. To one yeast RNA control, add 150 µL of RNase digestion buffer. To the other yeast control and the experimental samples, add 150 µL of the RNase solution. 3. Vortex, spin to collect, then incubate at 37°C for 30 min to digest unprotected RNA. 4. Add 225 µL inactivation/precipitation solution to each tube. Vortex and spin briefly. Precipitation of small fragments (100–150 nt) can be improved by adding 75 µL of ethanol, or 150 µL of ethanol for fragments 50–100 nt. Because for a typical 300-nt probe, many fragments may be less than 100 nt, we add 150 µL ethanol routinely. 5. Incubate at –20°C for 15 min. It is not necessary to add carrier RNA at this step.

3.4. Separation and Detection of Protected Fragments 1. Centrifuge the RNase-digested samples 15 min, at at least 10,000g, 4°C. 2. Carefully remove the supernatant from the tube, being careful not to dislodge the pellets. Spin again briefly and aspirate remainder of supernatant. A blue residue may be apparent at the bottom of the tube. This is a co-precipitant included in the RPA III kit that makes it easier to locate the pellet. This residue may not dissolve completely in the loading buffer.

Analysis of HPV Transcription by RPA

285

3. Resuspend pellets in 8 µL of gel loading buffer. 4. Heat the tubes to approx 90°C (±5°C) for 3–4 min to denature RNA and aid in solubilization, then revortex and pellet briefly. Also heat the size markers before loading onto the gel (see Note 13). 5. Load each sample on the gel and run at 150–300 V. Use only approx 1 µL of the undigested probe controls, as they are much more radioactive. 6. Transfer the gel to filter paper, cover with plastic wrap, and expose to film at –70°C with an intensifying screen. The length of exposure time will depend on the intensity of the signal, but overnight is a good starting place. Repeated exposures are possible, but once the gel has frozen, care should taken to prevent the gel from thawing while a new film is applied.

4. Notes 1. RPA analysis can be used to map promoter start sites or to measure the amount of mRNA originating from a particular start site (11). Because extraneous bands can appear in RPA analyses, initial characterization of promoters requires primer extension or S1/ExoVII analysis in addition to RPA to confirm that transcription initiates at a given location. Once the exact initiation sites are mapped with confidence, however, RPA can provide a convenient, simple, and sensitive way to measure transcription from a known promoter under different conditions. RPA analysis can also be used to measure expression of particular genes. One commonly performed RPA in our laboratory is to measure the level of L1 transcripts in raft cultures of HPV-infected keratinocytes as a step in determining whether a particular wild-type or mutant HPV can support late gene expression. Using this strategy, we can determine quantitatively the effect that, for example, a mutation in a transcriptional regulatory region has on expression of L1 late in the viral life cycle. A similar strategy can be used to study the expression of any of the other viral genes. If the probe is designed to overlap a spliced region, changes in splicing patterns and/or splice site usage under different conditions can be detected (3). RPA is one of a number of techniques for mapping and analyzing RNAs. Others include S1 and ExoVII nuclease mapping, primer extension, QRT-polymerase chain reaction (PCR), and Northern analysis. The specific information desired from the experiment will dictate which assay should be used in a particular case. S1/ExoVII mapping is similar in concept to RPA in that a single-stranded probe is hybridized to the sample, treated with nucleases, and analyzed by electrophoresis. The major conceptual difference with RPA is that S1/ExoVII probes are labeled at one end rather than uniformly throughout the length of the probe. This has several consequences. First, only fragments that include the labeled end of the probe will appear on the autoradiograph. This means that S1/ExoVII measures the distance between the labeled end and the first point of noncomplementarity with the target RNA. In contrast, fragments protected by any segment of the probe in an RPA can be used to obtain information about the target. Second, the signal intensity per mole of labeled probe is lower for S1/ ExoVII than for RPA. This results in a lower sensitivity of the assay. Third,

286

Bodily and Meyers

because only fragments containing the labeled end appear on the gel, the background is lower for S1/ExoVII assays than for RPA, in which extraneous bands can sometimes be present. The major advantage of S1/ExoVII analysis is that it is capable of unambiguously determining the distance between the end of the probe and the 5' end of an RNA molecule with low background. Primer extension analysis is capable of mapping the 5' ends of RNA molecules despite discontinuities, which can be both an advantage and a disadvantage. It is a disadvantage in that features such as splice sites that result in a shorter RNA cannot be unambiguously distinguished from an alternative transcriptional initiation site. In contrast, using RPA one can obtain information about the behavior of the target RNA at any point along the length of the probe. The advantage of primer extension is precision of mapping RNA ends. By performing a dideoxy sequencing reaction using the same primer and running the reaction products of the same gel, one can determine the precise nucleotide to which the 5' end maps. The mapping resolution of an RPA is limited by the accuracy of size comparison between the protected RNA fragments and size markers. Quantitative real-time (QRT)-PCR analysis allows the investigator to determine the absolute number of transcripts present in a sample, and therefore may be preferable to RPA for transcript quantitation in some situations. A significant difference between RPA and QRT-PCR is that, like primer extension, QRT-PCR is insensitive to sequence features, such as splice sites, that create discontinuities in the region to be analyzed, while RPA can detect such features. Whether this difference represents an advantage or disadvantage depends on the experimental situation. Northern blot analysis is useful for determining the size and quantity of RNA molecules containing the particular sequence complementary to a probe. In the context of HPV transcription, the multitude of transcriptional initiation sites, splices, and differentially polyadenylated transcripts can make interpretation of a Northern blot problematic. RPA can coalesce all of the species containing a particular sequence into a single signal, or alternatively, can distinguish features of different species, depending on the design of the probe. 2. A number of RNA isolation reagents are on the market. RNA from methylcellulose cultures can be extracted using protocols specified by the manufacturers for monolayer cultures after washing the suspended cells in phosphate-buffered saline (PBS) several times to eliminate the methylcellulose. Using TRIzol reagent (Invitrogen) to extract RNA from methylcellulose cultures, we have encountered the formation of aggregate material upon addition of TRIzol. This is lessened by the addition of a small amount of TRIzol to the cell pellet first, mixing well, and then adding more TRIzol gradually. RNA from raft cultures can be extracted using protocols for extraction from tissue samples, using, for example, glassTeflon® or power homogenizers to disrupt the tissue. Whatever the RNA isolation method, the RNA should be treated with RNase-free DNase to eliminate contaminating DNA from the preparation. This step is important to ensure that signals in the RPA result from protection of the probe by RNA and not by DNA


3.

4.

5.

6.

287

contaminants. The integrity of the RNAs should be confirmed by agarose gel electrophoresis and staining with ethidium bromide to observe the two prominent rRNA bands. Because the RNA and probe will be precipitated as part of the hybridization procedure, the concentration of RNA needed is not crucial: the precipitation can be adjusted to accommodate a wide range of volumes. We find 0.5–2 µg/µL total RNA to be a convenient range for a working concentration, and that 10–15 µg of RNA from HPV-infected cells in each assay reaction is enough to generate a robust signal from most HPV transcripts. Up to 50 µg of RNA can be used in the assay, but the sample may migrate aberrantly in the gel if too much RNA is present. Poly(A)-selected mRNA may also be used; 0.6–2 µg per reaction is a good starting place. One of the most persistent problems in working with RNA is avoiding contamination by ribonucleases. RNases are very stable and difficult to eliminate from a contaminated sample. Several helpful precautions are listed as follows. a. To avoid contamination by RNases from the hands, always wear gloves when working with RNA samples or with materials or reagents to be used in RNA work. Change gloves frequently. b. New, sterile, disposable plasticware, such as microcentrifuge tubes in a freshly opened bag, can be relied upon to be RNase free so long as they are handled so that RNases from the hands or other sources are not introduced. Glassware can be treated by baking at 180°C for 8 h or more. It is helpful to set aside materials and supplies to be used for RNA work and store them in a designated place so as not to mix them with other laboratory materials. c. Equipment such as pipetmen should be wiped down with RNase-inactivating solutions such as RNase Zap (Sigma). d. Solutions should be made using baked glassware or fresh plasticware, baked spatulas, and chemicals that are reserved for RNA work. Solutions can be treated with diethyl pyrocarbonate (DEPC) to eliminate RNase activity by adding DEPC to the solution to 0.1%, incubating for 12 h at 37°C, and autoclaving for 15 min at 15 lb/sq. in. (approx 1 bar or 100 kPa). Autoclaving inactivates the DEPC, so DEPC-treated solutions can subsequently become contaminated with RNases. Because DEPC reacts with amine groups, solutions containing Tris cannot be DEPC treated. Nuclease-free stock solutions of Tris can be obtained commercially and used to make working solutions. e. When manipulating solutions containing RNA, use aerosol barrier pipet tips. Several versions of the MAXIscript kit exist containing different combinations of phage polymerases. The polymerase needed to generate the riboprobe will dictate which kit to purchase (see Note 6). Make 10% APS fresh the day you use it. We keep 50-mg aliquots of powder to which we can add 500 µL of water to make a 10% solution, thereby having a fresh solution quickly without much waste. The riboprobe template consists of cloned DNA corresponding to the target RNA. This DNA may consist of genomic DNA or cDNA. Probe template design is

288

Bodily and Meyers

critical for extracting the most information from the RPA, and will depend on the needs of the particular experiment. The flexibility of the RPA permits a great deal of ingenuity in probe design. Probes can be made to overlap a splice junction (differentiating between spliced and unspliced populations of transcripts) or overlapping the 5' end of a transcript (permitting mapping of transcriptional start sites or differentiating between closely spaced promoters). One probe we have made includes sequences from a reporter cloned downstream of an HPV promoter. If the assay is performed on HPV-infected cells transfected with the reporter, RNA derived from the transfected reporter will protect a longer segment of the probe, thus permitting us to distinguish endogenous from transfection-derived transcripts. Using a probe template made from a spliced cDNA, one can probe specifically for a particular spliced product. If more than one protected fragment is expected from the digest, such as would be the case for two closely spaced promoters or perhaps spliced vs unspliced transcripts, the probe should be designed so that the resulting fragments are sufficiently different in size to be resolved on the gel. This is also true if more than one probe is used in the assay, such as when an internal control is included for quantitation. We have found high-fidelity PCR with primers flanking the region of interest followed by cloning of the fragment into a vector (such as pGEM) containing phage promoters on either side of the multiple cloning site to be a simple way to create any probe we desire. Whatever the probe design, there must be a phage promoter (such as T7, T3, or SP6) at the 3' end of the region of interest, oriented to initiate transcription into the region of interest. The distance between the phage promoter and the digested end of the linearized template will be the length of the undigested probe. Although probes of 1000 nt are possible, we have found that 200–450 nt is a convenient size of probe to work with, and the expected fragments protected by a probe this size are easily resolved by electrophoresis. The protected region of the probe should be sufficiently different in size from the full-length undigested probe to distinguish the two species. 7. There has been a report that transcription can occur from the opposite strand if the template is linearized with an enzyme that leaves a 3' overhang (12). The linearizing enzyme should therefore leave a blunt end or a 5' overhang. Alternatively, a 3' overhang can be filled in. 8. The reaction is assembled at room temperature to prevent the precipitation of spermidine present in the transcription buffer. The reaction volumes here are recommended as a starting place. The amount of probe needed will depend on the number of samples to be tested and the amount of probe to be added to each, and can be adjusted by adjusting the volumes of each component proportionally. We have found that a half reaction generates more than enough probe for a standard RPA of 12–20 samples. 9. Shorter elution times are possible with a corresponding decrease in recovery. Larger probes elute more slowly. We have found an overnight elution to be logistically convenient. Because of the isotopic label along its length, the probe will rapidly degrade by radiolysis, resulting in many extraneous bands, so probes


10.

11.

12.

13.

289

should be used within a few days of being made. We have obtained good results by setting up the hybridization the day after the probe was made (i.e., following the overnight elution step). Dissolving urea is facilitated by heating to 37°C and rocking the tube occasionally. Urea often contains particulate impurities that can cause blurring of bands when the gel is run, and boric acid in TBE buffers often precipitates. It is therefore helpful to filter the gel solution using a disposable filtration unit after the urea is dissolved, but before the addition of APS or TEMED. Leaving the vacuum on for a few min after the solution has passed through the unit also de-gasses the solution. We have also found it helpful after pouring the gel solution to place the comb only partway into the gel so that the well is shallow; the sample can become diffuse as it sinks to the bottom of a very deep well, resulting in fuzzy bands. Allow the gel to “cure” overnight at room temperature. Placing damp paper towels over the top of the gel and wrapping it in plastic wrap can prevent drying of the wells and precipitation of urea. When the RPA is used to quantitate the level of the target RNA, it is helpful to include a probe against an internal control transcript that is expressed at a constant level to provide a reference point against which the level of target RNA can be compared. It also provides a standard for normalization of RNA levels between samples. Although a number of “housekeeping” transcripts have been used in the literature as internal controls, the cyclophilin mRNA is most consistent over the course of keratinocyte differentiation (13). A template for making a probe against cyclophilin is available from Ambion. If an internal control is to be included, the probe against the control is made using the same protocol as the probe against the target RNA, and the control probe is added to each hybridization reaction along with the experimental probe. It is possible to add several experimental probes to the reaction to detect the levels of multiple transcripts at the same time, as is the strategy behind several commercially available systems for detecting multiple cytokine transcripts. It is important that the sizes of the protected fragments from each probe be sufficiently different to distinguish them on the gel. For example, we often measure the levels of HPV-31 L1 transcripts from cells grown in raft culture using a cyclophilin internal control. The protected fragment of the HPV31 L1 probe is 216 nt, while the cyclophilin fragment is 103 nt, sizes that are easy to tell apart using a 5% polyacrylamide gel. We usually hybridize overnight for convenience, so that the actual ribonuclease treatment and electrophoresis are performed 2 d after the probe was made. When we have tried shorter hybridization times, the results were not significantly altered. We use the RNA Century™ Markers from Ambion. The template for these markers consists of a mixture of linearized plasmid templates that generate a series of labeled in vitro transcripts using essentially the same protocol as described above for making the riboprobe. There is no need to gel-purify the markers. With freshly made markers, 1–2 µL of a 1:50 dilution is more than sufficient for a strong signal. We make dilutions directly in gel loading buffer and store them pre-diluted

290

Bodily and Meyers at –20°C. By using a greater volume or a lesser dilution, one marker reaction can be used for several months, until the signal intensity diminishes by radiolytic decay.

References 1. Meyers, C., Frattini, M. G., Hudson, J. B., and Laimins, L. A. (1992) Biosynthesis of human papillomavirus from a continuous cell line upon epithelial differentiation. Science 257, 971–973. 2. Ozbun, M. A. and Meyers, C. (1997) Characterization of late gene transcripts expressed during vegetative replication of human papillomavirus type 31b. J. Virol. 71, 5161–5172. 3. Ozbun, M. A. and Meyers, C. (1998) Human papillomavirus type 31b E1 and E2 transcript expression correlates with vegetative viral genome amplification. Virology 248, 218–230. 4. Ozbun, M. A. and Meyers, C. (1998) Temporal usage of multiple promoters during the life cycle of human papillomavirus type 31b. J Virol 72, 2715–2722. 5. Bechtold, V., Beard, P. and Raj, K. (2003) Human papillomavirus type 16 E2 protein has no effect on transcription from episomal viral DNA. J. Virol. 77, 2021–2028. 6. Frattini, M. G., Lim, H. B., and Laimins, L. A. (1996) In vitro synthesis of oncogenic human papillomaviruses requires episomal genomes for differentiationdependent late expression. Proc. Natl. Acad. Sci. USA 93, 3062–3067. 7. Sen, E., Bromberg-White, J. L., and Meyers, C. (2002) Genetic analysis of cis regulatory elements within the 5' region of the human papillomavirus type 31 upstream regulatory region during different stages of the viral life cycle. J. Virol. 76, 4798–4809. 8. Bromberg-White, J. L. and Meyers, C. (2003) Comparison of the basal and glucocorticoid-inducible activities of the upstream regulatory regions of HPV18 and HPV31 in multiple epithelial cell lines. Virology 306, 197–202. 9. Ruesch, M., Stubenrauch, F., and Laimins, L. (1998) Activation of papillomavirus late gene transcription and genome amplification upon differentiation in semisolid medium is coincident with expression of involucrin and transglutaminase but not keratin 10. J. Virol. 72, 5016–5024. 10. DiLorenzo, T. P. and Steinberg, B. M. (1995) Differential regulation of human papillomavirus type 6 and 11 early promoters in cultured cells derived from laryngeal papillomas. J. Virol. 69, 6865–6872. 11. del Mar Pena, L. M. and Laimins, L. A. (2001) Differentiation-dependent chromatin rearrangement coincides with activation of human papillomavirus type 31 late gene expression. J. Virol. 75, 10,005–10,013. 12. Schenborn, E. T. and Mierendorf, R. C., Jr. (1985) A novel transcription property of SP6 and T7 RNA polymerases: dependence on template structure. Nucleic Acids Res. 13, 6223–6236. 13. Steele, B. K., Meyers, C., and Ozbun, M. A. (2002) Variable expression of some “housekeeping” genes during human keratinocyte differentiation. Anal. Biochem. 307, 341–347.

Regulatory Motifs Within HPV Transcripts

291

22 Analysis of Regulatory Motifs Within HPV Transcripts Sarah A. Cumming and Sheila V. Graham Summary Papillomavirus late gene expression is highly dependent on host epithelial cell differentiation, such that capsid proteins are produced only in differentiating cells. Several papillomaviruses contain negative regulatory elements (NREs), that is, short regions of late transcripts that interact with host cellular RNA processing factors to prevent capsid protein synthesis in undifferentiated cells. In this chapter, the human papillomavirus (HPV)-16 NRE will be used as an example to show how cis-acting RNA regulatory elements can be identified and mapped using transient transfection of reporter gene constructs. The use of reporter gene assays is also readily applicable to the identification and characterization of novel promoters and other regulatory sequences in HPV DNA. In vitro RNA–protein binding techniques, including ultraviolet crosslinking, electrophoretic mobility shift assay, and affinity purification of RNA binding proteins, will also be described, again using the HPV-16 NRE as an example. These techniques may be used to identify cellular proteins that bind the NRE, allowing its mode of action to be deduced. They may also be used to study interactions between host cellular proteins and other protein-binding motifs on HPV mRNA. These interactions are important for the regulation of HPV gene expression, and have key roles in splicing, polyadenylation, mRNA export, stability, and translation.

1. Introduction Papillomavirus late gene expression is confined to the differentiated cells in the outer layers of an infected epithelium. In undifferentiated cells, although late mRNAs are present, capsid protein production may be prevented by negative regulatory elements (NREs). These are short RNA elements that have been identified in the open reading frames of the L1 and L2 late genes (1–4) or late 3' untranslated regions (UTRs) (5–8) of several papillomaviruses. NREs interact with host RNA processing factors and inhibit late gene expression by a variety of posttranscriptional mechanisms, including reduction of RNA stability (9), retention of transcripts in the nucleus (1,7), and reduction of polyadenylation efficiency (10). An NRE may be identified in papillomavirus From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

291

292

Cumming and Graham

DNA by cloning selected viral sequences into a mammalian reporter vector and assaying for reporter gene expression following transient transfection in basal epithelial cells. If expression is low compared to the control that lacks the viral sequences, an NRE is present. Once the limits of the element have been established by deletion analysis, NRE RNA may be made by in vitro transcription and incubated with epithelial cell extracts to identify NRE-binding proteins. If the RNA is radioactively labeled, the sizes of the proteins binding directly to the RNA may be determined by ultraviolet (UV) crosslinking the proteins to the RNA, sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE), and exposing the dried gel to X-ray film (11). Alternatively, the radiolabeled NRE RNA may be incubated with cellular proteins, then electrophoresed through a nondenaturing acrylamide gel in an electrophoretic mobility shift assay (EMSA) (12). Bound protein retards the migration of the probe through the gel, and incubation with antibodies against bound proteins can retard it still further, generating a supershift. It is also possible to purify NRE-binding proteins by chemically crosslinking unlabeled RNA to agarose beads, incubating with cellular extracts, and washing the beads with a high-salt buffer to remove proteins that are bound nonspecifically. The bound proteins are eluted in protein gel loading buffer and Western blotted to determine whether specific proteins are part of the complex that forms on the NRE. These techniques will be illustrated using the human papillomavirus (HPV)16 NRE as an example, but they may be more widely applied to the study of papillomavirus gene transcription (see Chapters 20 and 28), and of RNA processing events such as polyadenylation and alternative splicing. This chapter describes (1) selection of a reporter gene and plasmid design for experiments to identify and map an NRE; (2) transient transfection of reporter gene constructs into basal epithelial cells; (3) chloramphenicol acetyltransferase (CAT) reporter gene assays; (4) β-galactosidase reporter gene assays; (5) use of deletion constructs to map an NRE; (6) design, labeling, and purification of suitable probes to identify NRE-binding proteins; (7) UV crosslinking of NRE probes to proteins present in cellular extracts or purified expressed proteins; (8) EMSA using cellular extracts or purified proteins; (9) antibody supershift EMSA experiments; and (10) affinity purification and Western blotting of NRE-binding proteins present in cellular extracts. 2. Materials 2.1. Identification of an NRE 1. A suitable mammalian expression vector containing a strong promoter (e.g., the cytomegalovirus [CMV] immediate early [IE] gene promoter) and a reporter gene (e.g., CAT, lac Z, luciferase), either lacking polyadenylation signals, e.g., pLW1


2. 3. 4. 5.

293

(13), or from which polyadenylation signals may be readily removed and replaced if required, e.g., pGL3 or pCAT3 vectors (Promega). It may be advisable to use an intron-containing plasmid, e.g., pCI (Promega), to allow for possible interactions between splicing and polyadenylation factors across the terminal exon. Restriction endonucleases and 10X buffers as supplied by the manufacturer. Bacteria competent to receive plasmid DNA (stored at –70°C in small aliquots). T4 DNA ligase and reaction buffer as supplied by the manufacturer. Materials for polymerase chain reaction (PCR) amplification (see Subheading 2.11.).

2.2. Transient Transfection of Undifferentiated Epithelial Cells 1. 2. 3. 4.

Mammalian epithelial cells suitable for transient transfection, e.g., HeLa cells. Commercial transfection reagent (LipofectAmine from Invitrogen works well). 2 M CaCl2. Filter-sterilize and store at 4°C for up to 6 mo. 10X HEPES-buffered saline (HBS): 8.18% NaCl, 5.92% HEPES, 0.2% Na2HPO4 (pH 7.15). Filter-sterilize and store at 4°C for up to 6 mo.

2.3. CAT Reporter Gene Assay 1. CAT assay resuspension buffer: 110 mM Tris-HCl (pH 7.8), 1 mM ethylenediamine tetraacetic acid (EDTA), 1 mM dithiothreitol (DTT). Store at 4°C for up to 3 mo. 2. Xylene extraction buffer: 1 mM Tris-HCl (pH 7.5), 1 mM EDTA. 3. Chloramphenicol [ring-3, 5-3H] (NEN NET928) (see Note 1). 4. n-butyryl coenzyme A (C4:0) lithium salt (Sigma B1508). Store 25 mM stock in dH2O in small aliquots at –70°C for up to 3 mo. 5. 0.5 mM chloramphenicol. Make 50 mM stock in 100% ethanol, and then dilute 10 µL into 1 mL dH2O immediately before use. Store 50 mM stock at 4°C for no more than 2 d. 6. CAT enzyme from Escherichia coli 50,000–100,000 units per mg protein (Sigma C2900). Store the stock solution (0.1 ng/µL in dH2O) in small aliquots at –20°C for up to 3 mo. 7. Xylenes, mixed. ACS reagent ≥ 98.5% isomers plus ethylbenzene (highly flammable) (Sigma X2377). 8. Petroleum ether (highly flammable). 9. Liquid scintillation counter and scintillation fluid.

2.4. β-Galactosidase (ONPG) Assay 1. o-nitrophenyl-β-D-galactopyranoside (ONPG) substrate for β-galactosidase (Sigma N1127): 4 mg/mL in 0.1 M sodium phosphate (pH 7.5). Make fresh immediately before use. 2. 100X Mg solution: 0.1 M MgCl2, 4.5 M β-mercaptoethanol. 3. 0.1 M sodium phosphate buffer (pH 7.5). Store at room temperature. 4. 1 M Na2CO3. Store at room temperature.

294

Cumming and Graham

5. E. coli β-galactosidase enzyme purified from E. coli or recombinant: 3000 U/mL in 0.1 M sodium phosphate (pH 7.5). Store in small aliquots at –20°C.

2.5. Mapping an NRE 1. Materials for cloning (see Subheading 2.1.). 2. Materials for transient transfections (see Subheading 2.2.). 3. Materials for CAT reporter gene assays (see Subheading 2.3.) and β-galactosidase reporter gene assays (see Subheading 2.4.). 4. Materials for PCR (see Subheading 2.11.).

2.6. Probe Preparation 1. A plasmid containing a multiple cloning site (MCS) flanked by T3, T7, or Sp6 bacteriophage RNA polymerase promoter sequences, e.g., pBluescript (Stratagene), pGEM-T (Promega). 2. Materials for cloning (see Subheading 2.2.). 3. Materials for PCR (see Subheading 2.11.). 4. Diethylpyrocarbonate (DEPC, Sigma) (see Note 2). 5. [α-32P]-rUTP or -rCTP (800 mCi/mmol) and a safe environment for the use of radiochemicals. 6. T3, T7, or Sp6 RNA polymerase, dilution buffer, and 5X or 10X transcription buffer as supplied by the manufacturer. 7. Ribonucleotide triphosphate (NTP) set (100 mM). 8. DNase I enzyme that is free of RNase activity, e.g., RQ1 (Promega). 9. Sephadex G50 columns, e.g., Mini Quick spin RNA columns (Roche). 10. Equipment and reagents for denaturing PAGE (see Subheading 2.13.). 11. 0.5% SDS in TE buffer. 12. 3 M Sodium acetate (pH 5.2) treated with DEPC. 13. Phenol saturated with 0.1 M citrate buffer (pH 4.3) (e.g., Sigma P4682). Store at 4°C. 14. Chloroform (flammable). Store protected from light at room temperature. 15. 75% Ethanol, prepared using DEPC-treated dH2O. 16. Liquid scintillation counter and scintillation fluid.

2.7. UV Crosslinking 1. Equipment and reagents for SDS-PAGE (see Subheading 2.12.). 2. Diethylpyrocarbonate (DEPC, Sigma) (see Note 2). 3. HeLa cell nuclear extracts purchased from 4C Biotech (Seneffe, Belgium) or prepared according to Dignam (14). May be stored at –70°C for several months in single-use aliquots, which should be thawed on ice immediately before use. 4. Purified proteins expressed in bacteria, if available. 5. tRNA from E. coli strain W (Sigma R1753) or tRNA type X-SA from Bakers yeast (Sigma R8759). Prepare 10 mg/mL stock in RNase-free water (see Note 2) and store for 1 yr at –20°C.


295

6. 1X UV crosslinking binding buffer: 60 mM KCl, 20 mM HEPES-KOH (pH7.6), 1 mM MgCl2, 10% glycerol. Make 2X or 5X stock and store at room temperature for up to 3 mo. 7. Ribonuclease A (RNase A) type 1-A from bovine pancreas, 5X crystallized, salt fractionated, and chromatographically purified (Sigma R4875). Prepare 10 mg/mL stock in dH2O and store in small aliquots at –20°C for up to 1 yr. 8. 96-Well flat-bottomed plates. 9. Stratalinker® (Stratagene) or other suitable source of UV radiation at 254 nm. 10. Chromatography paper. 11. Gel dryer. 12. X-ray processor or chemicals, X-ray film, and a light-tight autoradiography cassette.

2.8. Electrophoretic Mobility Shift Assay 1. Equipment and reagents for nondenaturing polyacrylamide gel electrophoresis (see Subheading 2.13.). 2. Diethylpyrocarbonate (DEPC) (Sigma D5758) (see Note 2). 3. 1X EMSA binding buffer: 60 mM KCl, 10 mM HEPES-KOH (pH 7.6), 3 mM MgCl2, 5% glycerol, 1 mM DTT, 0.5 mM phenylmethylsulfonyl fluoride (PMSF). Make 5X stock (without PMSF or DTT) and store at room temperature for up to 3 mo. 4. Heparin: sodium salt grade 1-A, from porcine intestinal mucosa (Sigma H3393). Prepare 50 mg/mL stock in RNase-free water and store for up to 1 yr at –20°C. 5. DTT (dithiothreitol) molecular biology reagent (Sigma D9779). Prepare 1 M stock in 10 mM sodium acetate (pH 4.8) and store at –20°C for up to 1 yr. 6. HeLa cell nuclear extracts (see Subheading 2.7.3.). 7. Purified proteins expressed in bacteria, if available. 8. tRNA (see Subheading 2.7.5.). 9. RNase inhibitors (e.g., Promega RNasin®, N2111 or N2511) and protease inhibitors (e.g., PMSF, Sigma P7626. 0.2 M stock in isopropanol, store at –20°C, or Roche mini complete EDTA-free protease inhibitor cocktail 1836170). 10. Chromatography paper. 11. Gel dryer. 12. X-ray processor or chemicals, X-ray film, and a light-tight autoradiography cassette.

2.9. Antibody Supershifts 1. Reagents for EMSA (see Subheading 2.8.). 2. Antibodies against putative NRE-binding proteins.

2.10. Affinity Purification of RNA-Binding Proteins 1. Diethylpyrocarbonate (DEPC, Sigma) (see Note 2). 2. Templates and reagents for RNA in vitro transcription (see Subheading 2.6.). 3. Adipic acid dihydrazide agarose beads (Sigma).

296

Cumming and Graham

4. 14-mL Round bottomed polypropylene centrifuge tubes (Greiner). 5. 1 M Sodium acetate, pH 5.0. 6. Sodium m-periodate (Sigma). Store 50 mM stock at room temperature (protected from the light) for up to 3 mo. 7. Polyuridylic acid agarose (Sigma). Immediately before use, add 100 mg beads to 1 mL 2 M NaCl and leave on ice to swell for 1 h. 8. 2 M NaCl prepared using DEPC-treated dH2O. 9. HeLa cell nuclear extracts (see Subheading 2.7.3.). 10. High-salt buffer D: 20 mM HEPES-KOH (pH 7.6), 5% glycerol, 300 mM KCl, 0.2 mM EDTA, 0.5 mM DTT (prepare fresh or store without DTT at room temperature for up to 3 mo). 11. Low-salt buffer D: 20 mM HEPES-KOH (pH 7.6), 5% glycerol, 100 mM KCl, 0.2 mM EDTA, 4 mM MgCl2, 0.5 mM DTT (prepare fresh or store at room temperature without DTT for up to 3 mo). 12. Equipment and reagents for SDS-PAGE (see Subheading 2.12.). 13. Equipment (electroblotting cell) and membrane (e.g., Hybond P, Amersham Pharmacia Biotech) for Western blotting. 14. Chromatography paper. 15. Western blotting transfer buffer: 25 mM Tris (pH 8.3), 192 mM glycine, 20% methanol. 16. Primary antibodies against RNA binding proteins. 17. Horseradish peroxidase (HRP)-conjugated secondary antibodies. 18. Dried skimmed-milk powder. 13. PBS (140 mM NaCl, 2.7 mM KCl, 10 mM Na2HPO4, 1.8 mM KH2PO4, adjust pH to 7.4 using HCl, store at room temperature). 14. Polyoxyethylene-sorbitan monolaurate (Tween-20, Sigma P7949). 15. PBS-T (PBS + 0.05% [v/v] Tween-20). 16. Western blot detection reagent, e.g., ECL (Amersham Pharmacia Biotech). 17. X-ray processor or chemicals, X-ray film, and a light-tight autoradiography cassette. 18. Western blot stripping buffer: 100 mM β-mercaptoethanol, 2% SDS, 62.5 mM Tris-HCl (pH 6.8). Prepare immediately before use, adding the mercaptoethanol in the fume hood.

2.11. Reagents for PCR Amplification 1. 2. 3. 4. 5.

Thermal cycler. dNTP mix (10 mM each, store at –20°C). Taq polymerase and 10X PCR buffer supplied by the manufacturer. 25 mM MgCl2. Custom oligonucleotide primers, resuspended in sterile dH2O at 100 pmol/µL and stored at –20°C (10X stock).


297

2.12. Materials for SDS-PAGE 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

Small gel electrophoresis kit for SDS-PAGE, e.g., BioRad mini Protean system. Long round tips for loading gels, e.g., Sorenson Multiflex (13790). 30% Acrylamide solution (acryl/bis 37.5:1) (store at 4°C). TEMED (store at 4°C). 10% Ammonium persulphate (APS), store at 4°C for up to 1 mo. 1.5 M Tris-HCl (pH 8.8). 1 M Tris-HCl (pH 6.8). 10% SDS. Protein gel electrophoresis buffer: 0.25 M glycine, 0.025 M Tris, 0.5% SDS. 2X protein loading buffer: 100 mM Tris-HCl (pH 6.8), 4% SDS, 20% glycerol, 0.2% bromophenol blue, 200 mM DTT. Store at room temperature, adding DTT immediately before use. 11. DTT molecular biology reagent (Sigma). Prepare 1 M stock in 10 mM sodium acetate (pH 4.8) and store at –20°C for up to 1 yr.

2.13. Materials for Denaturing and Nondenaturing PAGE 1. 2. 3. 4. 5. 6. 7. 8.

Equipment for large polyacrylamide gel, e.g., BioRad Protean gel system. Long round tips for loading gels, e.g., Sorenson Multiflex (13790). 40% Acrylamide solution (acryl/bis 19:1) (for denaturing gel—store at 4°C). Urea (for denaturing gel). 30% Acrylamide solution (acryl/bis 37.5:1) (store at 4°C). TEMED (store at 4°C). 10% Ammonium persulphate (APS); store at 4°C for up to 1 mo. RNA loading buffer: 15% Ficoll, 1 mM EDTA, 0.25% bromophenol blue, 10 mM sodium phosphate (pH 7.0). Store at room temperature in small batches. 9. 10X TBE buffer: 0.9 M Tris, 0.9 M borate, 25 mM EDTA.

3. Methods 3.1. Identification of an NRE Functional assays should be carried out to show that the putative NRE sequence inhibits gene expression in undifferentiated epithelial cells, and to identify specific sequences within the NRE that are required for inhibition. Before starting these experiments, a reporter gene system must first be selected. Suitable choices include the E. coli CAT gene, the E. coli β-galactosidase gene, or firefly luciferase genes. CAT expression is relatively stable in transiently transfected cells and simple to assay using a phase-extraction method (15). It requires the use of some expensive reagents—e.g., tritiated chloramphenicol and butyryl coenzyme A. Access to a liquid scintillation counter is also a requirement. The β-galactosidase assay is inexpensive and

298

Cumming and Graham

simple to carry out, though in certain cell lines endogenous enzymes may give high background. It is conveniently used as a transfection control in combination with CAT assays using a portion of the same cell lysates, and requires only a spectrophotometer to measure samples. CAT and β-galactosidase assays will be described under Subheadings 3.3. and 3.4., respectively. The luciferase system is much more sensitive than the CAT reporter (16), inexpensive, and quick and simple to use, although it requires the use of a luminometer, which may not be readily available. Renilla (sea pansy) and Photinus (firefly) luciferase genes may be assayed from the same samples, providing an internal transfection control. A good range of plasmid vectors, reagents, and kits for assaying luciferase, CAT, and β-galactosidase is available from Promega. If the sequence of interest forms part of the 3' UTR of the virus (1,6,8,17), clone it downstream of a reporter gene in a mammalian expression vector that contains a strong promoter. If possible, select a plasmid that has an intron upstream of the reporter gene, since interactions between proteins bound to the NRE and to upstream splice sites may occur. The plasmid must include polyadenylation signals, i.e., a GU/U-rich CstF-64 binding site (18–20) downstream of a poly(A) hexanucleotide signal (AAUAAA). These may be the endogenous viral sequences, or the heterologous sequences present in the reporter plasmid. It may be necessary to test both, as the NRE may bind proteins that interact with specific downstream polyadenylation signals. If the sequence of interest is part of the coding region, e.g., the HPV-16 L1 and L2 gene inhibitory elements, the whole open reading frame may be cloned between the reporter gene and the polyadenylation signals, and the inhibitory elements mapped by deletion analysis (1,2). A second possibility is to generate an in-frame fusion between the putative inhibitory element and heterogeneous sequences that can be assayed at RNA or protein level (21), e.g., the CAT reporter gene. Alternatively, putative inhibitory elements may be analyzed in their natural position within the gene by introducing mutations that disrupt the nucleotide sequence while maintaining the amino acid sequence of the protein, allowing the protein produced to be quantified by Western blotting (21). The latter two approaches are preferable, since the position of the inhibitory element within the RNA might affect its function.

3.2. Transfection of Undifferentiated Epithelial Cells 1. Transient transfection experiments should be performed using undifferentiated epithelial cells, e.g., HeLa (which may be considered a convenient model for basal epithelial cells). Controls for the experiment should include mock-transfected cells, a plasmid known to express the reporter gene, and a plasmid that lacks the reporter gene. The cells should be co-transfected with a second plasmid expressing β-galactosidase (e.g., pSVβgal, Promega), to correct for any variations in transfection efficiency.


299

2. Plate the cells at 1 × 105 per 35-mm dish 16–24 h before transfection. 3. In a sterile tube, pipet 187 µL filter-sterilized 2X HBS, pH 7.15. The pH of this reagent is critically important. 4. In a second sterile tube, mix 150 µL sterile dH2O, 23 µL 2 M CaCl2, 3 µg test plasmid DNA, and 3 µg pSVβgal DNA. Note: This quantity is sufficient for three 35-mm dishes so the transfection may be performed in triplicate. 5. Slowly mix the contents of the two tubes, which should now look only very slightly cloudy. Incubate at room temperature for 30 min. If a coarse precipitate forms, check the pH of your HBS. 6. Pipet the transfection mixture slowly into the cell-culture medium. Change the medium after 16 h. Alternatively, transfect the cells using a commercial reagent (e.g., LipofectAmine, Invitrogen) in accordance with the manufacturer’s instructions.

3.3. CAT Reporter Gene Assay 1. After a further 24 h, scrape cells and medium into a centrifuge tube. 2. Centrifuge at 60g in a tabletop centrifuge for 5 min at 4°C. Decant the supernatant (medium). 3. Resuspend the cells in 1 mL ice-cold PBS, and transfer to a 1.5-mL microcentrifuge tube. 4. Centrifuge at full speed in a microcentrifuge for 1 min at 4°C, then carefully pipet off and discard the supernatant. 5. Resuspend the cells in 200 µL ice-cold resuspension buffer. 6. Transfer the samples to an ethanol/dry-ice bath for 2 min to freeze, then a 37°C water bath for 2 min to thaw. Repeat the freeze–thaw twice more. 7. Add 1 µL (0.1 ng) CAT enzyme to a sample of mock-transfected cells. Repeat the freeze-thaw once more. 8. Chill the samples on ice, and then centrifuge for 3 min at full speed at 4°C in a refrigerated microcentrifuge. 9. Transfer the supernatants to fresh tubes. Remove 30 µL from each and retain on ice for the β-galactosidase assay. 10. Heat the samples to 65°C for 6 min. This inactivates cellular enzymes that can use coenzyme A derivatives as substrates. It is vital to remove the lysate for the β-galactosidase assay before heating, because β-galactosidase is also inactivated at this temperature. 11. Chill the samples on ice. Add to each sample 20 µL 0.5 mM chloramphenicol, 2 µL [3H]-chloramphenicol, and 2 µL 25 mM butyryl coenzyme A. Incubate the reactions at 37°C for 2 h. The CAT enzyme transfers the butyryl groups from butyryl coenzyme A to chloramphenicol. Butyrylated chloramphenicol is soluble in xylene, allowing it to be separated from unmodified chloramphenicol (15). 12. Working in a fume hood, add 400 µL xylene, vortex thoroughly, and centrifuge at top speed for 2 min in a microcentrifuge. 13. In a fume hood, transfer the top (xylene) layer to a clean tube. 14. Add 200 µL xylene extraction buffer, vortex, centrifuge at top speed for 2 min, and then transfer the top layer to a clean tube. Repeat twice more.

300

Cumming and Graham

15. Add 30 µL from the purified sample to 2 mL scintillation fluid and count using a liquid scintillation counter (see Note 3).

3.4. β-Galactosidase (ONPG) Assay 1. For each sample, mix together 3 µL 100X Mg solution, 66 µL ONPG, 30 µL cell lysate, and 201 µL 0.1 M NaPO4 (pH 7.5). Also include a sample containing lysate from mock-transfected cells to estimate endogenous enzyme. For the positive control for the assay, add 1 µL of a 1 in 60 dilution of β-galactosidase enzyme to mock-transfected cells. Incubate the reactions at 37°C for 30 min or until a yellow color develops. ONPG is catalyzed by β-galactosidase in the presence of ATP to form the yellow compound o-nitrophenyl (22). 2. Measure the optical density of the samples at 420 nm (OD420) using a spectrophotometer. The readings obtained from the transfected cell lysates should all be very similar, if the transfection efficiency is consistent between samples. 3. Calculate the mean of the OD420 readings. Divide each OD420 by the mean, and multiply by 1 to obtain a series of values normalized to the mean of the data (e.g., 1.05, 1.10, 1.12, and so on). Correct the CAT assay scintillation counts for variation in transfection efficiency by dividing each by its corresponding normalized β-galactosidase value. If the plasmid containing the viral DNA sequences has low expression of the CAT reporter gene compared to the positive control, despite containing an adequate polyadenylation signal, the existence of an NRE can be deduced.

3.5. Mapping an NRE 1. The inhibitory element may be mapped within the 3' UTR or coding sequences by using PCR to amplify fragments that may contain the NRE. The PCR primers should each be designed to include as a 5' extension, restriction sites that will allow cloning of the PCR product into the reporter plasmid. Fig. 1A shows some of the plasmid constructs that were used to map the HPV-16 negative regulatory element. Progressive deletions were made from the 5' end of the late gene 3' UTR, until the loss of inhibitory sequences caused an increase in CAT reporter gene activity (Fig. 1B), allowing the precise location of the 5' end of the inhibitory element to be deduced (5). The 3' end of the element was subsequently identified using a complementary series of deletions from the 3' end of the late 3' UTR (23). 2. The position of the NRE should be confirmed by precise deletion, in order to locate the 5' and 3' ends of the element. For the HPV-16 NRE, removal of the 79-nt element leaving the surrounding sequence intact gave high gene expression comparable to that obtained using ∆G in Fig. 1B (24). The position of the NRE was further confirmed by using plasmid constructs that contain deletions within the NRE (Fig. 1C). Deletion of each region of the NRE tested led to an increase in CAT reporter gene activity, though gene expression was still significantly inhibited, showing that the whole 79-nt element is required for the most efficient inhibition of gene expression (24) (Fig. 1D). Deletions of portions of the NRE were generated using PCR (see Note 4, Fig. 2).


301

Fig. 1. Identification of an NRE in the HPV-16 late 3' untranslated region (UTR) using chloramphenicol acetyltransferase (CAT) reporter gene assays. (A) Diagram of plasmid constructs containing 5' end deletions that were used to map the HPV-16 late gene NRE. The positive control is a plasmid, pTer5, which contains the HSV-2 IE gene 5 mRNA 3'-end processing signals downstream of the CAT reporter gene (13). ∆A-∆G are plasmids that contain progressive 5'-end deletions of the late 3' UTR downstream of the CAT reporter. (B) Bar chart showing CAT activity assayed in the presence of [3H] chloramphenicol of HeLa cells transiently transfected with the CAT reporter plasmid constructs ∆A-∆G. (C) Diagram of plasmid constructs containing internal deletions of the NRE. The sizes of the wild-type and mutant NREs are: wt (79 nt); mut 1 (49 nt); mut 2 (30 nt); mut 3 (63 nt); and mut 4 (46 nt). (D) Bar chart showing CAT activity assayed in the presence of [3H]-chloramphenicol of HeLa cells transiently transfected with CAT reporter plasmid constructs containing deletions within the NRE. The positive control in this experiment, plasmid ∆G, lacks the NRE and upstream sequences and is shown in A. pBS (Stratagene) contains no CAT reporter gene. Arrow, HSV-2 immediate early (IE) gene -4/-5 promoter; open box, CAT reporter gene; stippled box, NRE; arrowhead, poly(A) site. 3. The NRE sequence should be examined to identify binding sites for RNA processing factors or other proteins, which might mediate the inhibitory activity of the element. Possible sequences of interest include 5' splice sites, which are found

302

Cumming and Graham

Fig. 2. Diagram showing how polymerase chain reaction (PCR) can be used to introduce site-specific mutations (see Note 4). (A) Template DNA showing the positions of PCR primers and restriction sites. (B) PCR amplification of two half molecules. (C) Products from the first round of PCR. (D) Annealing and PCR amplification of the two half molecules. (E) PCR product containing the desired mutation. Arrows, PCR primers; asterisk, mismatch generating desired mutation. in the bovine papillomavirus (BPV)-1 and HPV-16 NREs and bind U1 snRNP (24,25); AU-rich elements (AREs), e.g., in the HPV-1 NRE, which bind HuR (9); and GU-rich regions of the HPV-16 and -31 NREs, which bind CstF-64 (8,23). These may be identified by comparison with consensus splicing sequences (26) or GU-rich downstream sequence elements (18–20), or by consulting the AU-rich element website (http://rc.kfshrc.edu.sa/ared/). These may then be mutated by PCR (27) to determine what contribution such short sequences make to the inhibitory capacity of the element. In some cases, mutations affecting specific motifs will reduce the efficacy of the NRE while simultaneously reducing binding of specific proteins or complexes (9,10,24,28). This may suggest the mechanism by which the NRE inhibits gene expression in undifferentiated cells.


303

3.6. Probe Preparation The NREs that have been described to date bind splicing, polyadenylation, and RNA export and stability factors. Individual RNA–protein interactions may be demonstrated by preparing in vitro transcribed sense-strand RNA probes, which may be used in protein-binding experiments. 1. Clone the NRE into a vector containing a bacteriophage T3, T7, or Sp6 RNA polymerase promoter. If the sequence exceeds 200 bp, clone it in shorter portions to improve probe labeling and resolution of RNA–protein complexes. Linearize cesium chloride-purified plasmid DNA (22) downstream of the NRE sequence, then phenol/chloroform extract and ethanol precipitate. Enzymes that generate 3' overhangs must be avoided, because back transcription may occur, producing a mix of sense and antisense molecules. Preferably, since we have found that proteins present in cellular extracts may bind to certain plasmid polylinker sequences, prepare probe template by PCR amplification using a 5' primer that contains at its 5' terminus the T3 promoter sequence (5' 6 random nucleotides + AATTAACCCTCACTAAAGGG + 18–20 specific nucleotides) together with a conventional reverse primer. Gel purify the PCR product and resuspend it at 0.5–1 mg/mL in TE buffer. 2. All solutions for probe labeling, UV crosslinking, EMSA, and affinity chromatography must be RNase free (see Note 2). 3. Label the probe using [α-32P] rUTP or rCTP, depending on which occurs more frequently in the sequence of interest. rUTP is preferable, since rCTP may be modified to dCTP when incubated with cellular extracts. 4. Add to a microcentrifuge tube 0.5–1 µg template DNA; 2 µL NTP mix (2.5 mM each GTP, CTP, and ATP for UTP probes); 2 µL 5X buffer (supplied with the enzyme); 1 µL 100 µM DTT; 1.2 µL 100 µM UTP, 5 U T3, T7, or Sp6 polymerase; 20 U RNase inhibitor (optional); and 2.5 µL [α-32P] UTP. Dilute the polymerases immediately before use in the dilution buffer supplied with the enzyme, since these enzymes are highly thermolabile. The dilution buffer must be stored at –20°C and transferred from the freezer to an ice block only immediately before use to prevent it from warming up. Incubate the transcription reaction at 37°C for 1 h. 5. To remove the template, add 1 U DNase I enzyme and the supplied reaction buffer to 1X final concentration; incubate at 37°C for 1 h. 6. For UV crosslinking, a low level of transcripts that are not full length may be tolerated, though it is advisable to check a portion of the transcription reaction on a gel as described under Subheading 3.6.7. If the transcription reaction is satisfactory unincorporated, nucleotides may be removed using a Sephadex G50 spin column. If short transcripts are common, it is preferable to gel purify UV crosslinking probes (see Subheading 3.6.7. and Note 5), resuspending the probe RNA at 5 × 105 cpm/µL in DEPC-treated dH2O. 7. For EMSA, even a small proportion of transcripts that are not full length cannot be tolerated, since each may bind proteins and be retarded to a different extent.

304

Cumming and Graham Purify full-length transcripts from a 20-cm denaturing acrylamide gel (4% acrylamide, 50% w/v urea, 1X TBE; add APS to 0.001% and TEMED to 0.001% to polymerize). Add 3 µL RNA loading dye to the in vitro transcription reaction prepared as described under Subheading 3.6.4. Rinse the wells thoroughly before loading the samples. Electrophorese the samples in 1X TBE buffer at 10 V/cm for 2 h. Excise full-length transcripts from the gel (see Note 5), then elute the RNA from the gel slice at 37°C overnight in 0.5% SDS in TE buffer. Remove and discard the gel slice, then purify the RNA by phenol/chloroform extraction (using phenol buffered with citrate buffer, pH 4.3) and ethanol precipitation. Recover the RNA by centrifugation, wash the pellet with 75% ethanol, and then resuspend it in 50 µL DEPC-treated dH2O. Count using a scintillation counter, and dilute further if necessary to 5 × 104 cpm/µL.

3.7. UV Crosslinking This technique is used to determine the approximate sizes of proteins that can bind an NRE. A radiolabeled probe is incubated with nuclear or whole-cell extracts, or purified proteins. Proteins in direct contact with the RNA are covalently linked using UV irradiation. Unbound probe is digested using RNase enzyme, and then the radiolabeled bound proteins are separated by SDS-PAGE. Once approximate sizes have been determined, potential binding proteins of these sizes may be tested using EMSA supershift experiments (Subheadings 3.8. and 3.9.) or affinity purification and Western blotting (Subheading 3.10.). For example, the HPV-16 NRE binds proteins of approx 65 kDa that include the splicing factor U2AF65, and the polyadenylation factor CstF-64. It also binds HuR, a protein involved in RNA transport and stability. Binding was shown by UV crosslinking of purified protein and EMSA antibody supershift experiments (29). Fig. 3 shows HeLa cell nuclear extracts and glutathione S-transferase (GST)-tagged HuR protein UV crosslinked to the HPV-16 NRE. 1. Prepare a 12% SDS-PAGE minigel. Separating gel: 12% acrylamide, 0.375 M Tris (pH 8.8), 0.001% SDS; add APS to 0.001% and TEMED to 0.001% to polymerize. Stacking gel: 5% acrylamide, 125 mM Tris (pH 6.8), 0.001% SDS; add APS to 0.001% and TEMED to 0.001% to polymerize. 2. In a 96-well plate, mix together at room temperature in 1X UV crosslinking binding buffer, probe (5 × 105–1 × 106 cpm), 20 µg E. coli tRNA, and 20 µg nuclear extract or 0.5 µg purified expressed protein. The total reaction volume should be 20 µL. Controls for UV crosslinking experiments include antisense probes, probe with no nuclear extract (to ensure complete RNase digestion of the probe, since undigested probe fragments can easily be mistaken for small bound proteins), and, when GST-tagged purified proteins are used, GST alone. 3. Incubate the reactions at room temperature for 15 min. 4. UV crosslink on ice (first removing the lid) at 250 mJ using a Stratalinker.


305

Fig. 3. Ultraviolet crosslinking of human papillomavirus (HPV)-16 negative regulatory element (NRE) probes to HeLa cell nuclear extracts and glutathione S-transferase (GST)-tagged HuR protein. HPV-16 NRE and control probes were [32P]-labeled and crosslinked to 20 µg HeLa cell nuclear extracts (NE) (lanes 1–3), 0.5 µg GSTtagged HuR protein (lanes 5 and 6), or 0.5 µg GST protein. Lane 1, NRE (79 nt) + NE; lane 2, antisense NRE (79 nt) + NE; lane 3, in vitro transcribed pBS polylinker RNA (65 nt) + NE; lane 4, NRE probe only; lane 5, NRE + GST-HuR; lane 6, antisense NRE + GST-HuR; lane 7, NRE + GST protein. 5. Add 20 µg RNase A per reaction, place the 96-well plate in a Perspex box, and incubate at 37°C for 30 min. RNase-free tips and solutions are not required from this point. 6. Add 20 µL 2X protein loading buffer (containing 100 mM DTT) and transfer the samples to screw-capped tubes. 7. Denature the samples at 90°C for 10 min; load half the sample (20 µL) and separate by SDS-PAGE using 1X protein gel electrophoresis buffer. It is advisable to ensure all the bromophenol blue is run off the bottom of a UV crosslinking gel, since this will ensure the digested excess probe is also lost from the gel. 8. Dry the gel and expose it to X-ray film overnight (see Note 6).

306

Cumming and Graham

Fig. 4. Electrophoretic mobility shift assay (EMSA) competition and salt titration assays using a human papillomavirus (HPV)-16 negative regulatory element (NRE) probe. (A) A [32P]-labeled NRE probe (1.5 pmol) was incubated with HeLa cell nuclear extracts in the presence of specific competitor RNA. Lane 1, no extract; lane 2, no competitor; lanes 3–8, 1- to 16-fold molar excess of specific competitor, i.e., 1.5–24 pmol in vitro transcribed unlabeled NRE RNA. Arrows, RNA–protein complexes. NE, HeLa cell nuclear extracts. (B) A [32P]-labeled NRE probe (1.5 pmol) was incubated with HeLa cell nuclear extracts in the presence of varying concentrations of KCl. Lane 1, no extract; lane 2, 60 mM KCl; lane 3, 120 mM KCl; lane 4, 250 mM KCl; lane 5, 500 mM KCl. Arrows, RNA–protein complexes. NE, HeLa cell nuclear extracts.

3.8. Electrophoretic Mobility Shift Assay (EMSA) This method also involves binding radiolabeled NRE RNA probes to proteins present in cellular extracts, but the reactions are separated using a nondenaturing polyacrylamide gel, such that protein–protein as well as RNA– protein interactions are maintained. Complexes that form upon the RNA may contain several different proteins, and cause the RNA probe to migrate more slowly through the gel. Experiments using competitor RNAs may be used to demonstrate that the RNA–protein interactions are specific (see Notes 7 and 24). Figure 4A shows a competition experiment using unlabeled NRE RNA, which competes with an NRE probe to bind proteins present in HeLa cell nuclear extracts. Figure 4B shows the effect of increasing salt concentration on the binding of proteins in HeLa cell nuclear extracts to an NRE probe.


307

1. Prepare a 5% polyacrylamide gel, using 0.5-mm spacers and comb and 20-cm plates (5% acrylamide, 0.5X TBE, add APS to 0.01% and TEMED to 0.001% to polymerize). After polymerization, pre-run the gel at 200 V (10 V/cm) at 4°C in 0.5X TBE. 2. Assemble in microcentrifuge tubes on ice, a reaction containing 1X EMSA binding buffer, 5 × 104 cpm probe, 1 µg E. coli tRNA, and 20 µg nuclear extract or 0.5 µg purified protein in a total volume of 20 µL. It is important to keep the volume of the protein to less than 2 µL to maintain the salt concentration in the reaction. For each probe, include as a control one reaction that contains no proteins. 3. Incubate the reactions on ice for 15 min. 4. Add 1 µg heparin to each tube to stabilize the complexes. 5. Incubate on ice for 15 min. 6. Using long tips, load the samples without adding loading dye (see Note 8). 7. For a probe of around 100 nt, run the gel for 2–3 h at 200 V (10 V/cm). For longer probes, use a lower-percentage gel and a longer run time (e.g., 4% acrylamide in 0.5X TBE for 3.5 h for a 300-nt probe). Dry the gel and expose to film.

3.9. Antibody Supershifts To show that a known protein is part of an RNA–protein complex, an antibody against this protein may be added to the binding reaction. If the protein is present, the mobility of the complex is further reduced, generating a supershift (see Note 9). Alternatively, if binding of the antibody to its epitope interferes with RNA binding, the intensity of the complex may be reduced. Figure 5A, lane 2 shows that GST-tagged HuR protein can directly bind the HPV-16 NRE, and that addition of the anti-HuR monoclonal antibody 19F12 further reduces the mobility of the complexes, producing a supershift (lane 3). 1. Prepare and pre-run the gel as described under Subheading 3.8.1. 2. In a microcentrifuge tube on ice, mix together 20 µg nuclear extract, 1 µL (1 µg) antibody, and 1 µg E. coli tRNA in 1X binding buffer (see Note 10) in a (20-x) µL reaction, where x is the volume of probe required to give 5 × 104 cpm. Do not add the probe yet. 3. Incubate the samples on ice for 15 min. Add the probe. 4. Continue from Subheading 4.4.3. onwards.

3.10. Affinity Purification of NRE Binding Proteins In vitro transcribed RNA may be chemically linked to agarose beads, then incubated with cellular extracts to purify RNA-binding proteins (30). NREbinding proteins are eluted from the agarose beads and Western blotted. This technique was used to show that the U1 snRNP components U1A and Sm proteins bind to the HPV-16 NRE (24). Fig. 6 shows that HuR protein binds to the NRE and to poly(U) RNA, but not to agarose beads alone.

308

Cumming and Graham

Fig. 5. Electrophoretic mobility shift assay (EMSA) antibody supershift experiment using a human papillomavirus (HPV)-16 negative regulatory element (NRE) probe. EMSA experiment showing binding of [32P]-labeled NRE probes to 0.5 µg glutathione S-transferase (GST)-tagged HuR protein, and supershift of the RNA–protein complexes using 200 ng of the anti-HuR monoclonal antibody 19F12 (Molecular Probes). Lane 1, NRE probe; lane 2, NRE probe + GST-HuR; lane 3, NRE probe + GST-HuR + 19F12. Arrows, GST-HuR bound to NRE RNA; arrowheads, supershifted RNA–protein complexes.

Fig. 6. Western blot of proteins affinity purified on negative regulatory element (NRE) RNAs linked to agarose beads. Proteins were purified from HeLa cell nuclear extracts using RNA chemically crosslinked to adipic acid dihydrazide agarose beads and Western blotted using the anti-HuR monoclonal antibody 19F12 (Molecular Probes). Lane 1, NRE RNA linked to beads incubated with HeLa cell nuclear extracts (NE); lane 2, poly(U) RNA linked to beads incubated with NE; lane 3, beads incubated with NE; lane 4, NE only.


309

1. Prepare NRE RNA by in vitro transcription essentially as described under Subheading 3.6.4., substituting an NTP mix containing all four NTPs at 2.5 mM each, and omitting the unlabeled and radiolabeled UTP. Make the reaction volume up to 10 µL with DEPC-treated dH2O. Incubate the reaction at 37°C for 1 h, then DNase I digest the reaction as described under Subheading 3.6.5. Recover the RNA by phenol/chloroform extraction and ethanol precipitation. After recovering the RNA by centrifugation, wash with 75% ethanol and resuspend it in 30 µL DEPC-treated dH2O. Determine the concentration of the RNA using a spectrophotometer (1 OD260 unit = 40 µg/mL RNA) and check the integrity of the transcripts by gel electrophoresis. 2. Oxidize 500 pmol RNA in a 500 µL reaction containing 5 mM sodium m-periodate and 0.1 M sodium acetate (pH 5.0) at room temperature for 1 h in the dark (e.g., in a lead pot). Ethanol precipitate the RNA and resuspend it in 500 µL 0.1 M sodium acetate (pH 5.0). 3. Prepare adipic acid dyhydrazide agarose beads. Prepare beads also for a no RNA control. Add 10 mL 0.1 M sodium acetate to 400 µL beads in a 14-mL roundbottomed centrifuge tube. Spin the beads at 20–30g in a tabletop centrifuge for 3 min at 4°C. Carefully pipet off and discard the supernatant. Wash the beads three more times with 10 mL 0.1 M sodium acetate and resuspend them in 300 µL 0.1 M sodium acetate. 4. Add the beads to the oxidized RNA in a microcentrifuge tube and incubate the samples at 4°C overnight on a rotating wheel. 5. If appropriate, poly(U) agarose (polyuridylic acid agarose) beads may be used as a positive control for protein binding; e.g., poly(U) binds RNA-processing factors such as U2AF65, HuR, that also bind the 3' end of the HPV-16 NRE. Prepare the beads 1 h before use as described under Subheading 2., then wash the beads as described under Subheading 3.10.6., before incubation with cellular proteins. 6. Spin the beads at 20–30g at 4°C for 3 min. Pipet off and discard the supernatant. Gently resuspend the beads in 1 mL ice-cold 2 M NaCl and spin again. Wash the beads twice more with 1 mL 2 M NaCl. Add 1 mL high-salt buffer D to the beads and centrifuge at 20–30g for 3 min at 4°C. Discard the supernatant and wash three times more with 1 mL high-salt buffer D. 7. Resuspend the beads in 400 µL high-salt buffer D. Add 250 µL nuclear, cytoplasmic, or cellular protein extracts (250 µg) (see Note 11). Incubate at 30°C for 20 min, gently mixing occasionally. 8. Centrifuge the sample at 20–30g for 3 min at 4°C. Resuspend the beads in 1 mL low-salt buffer D and transfer to a 14-mL centrifuge tube. Add a further 10 mL of low-salt buffer D, mix gently, and centrifuge in a tabletop centrifuge as before. Wash twice more with 10 mL low-salt buffer D. Resuspend the beads in 1 mL low-salt buffer D, transfer to a microcentrifuge tube, and centrifuge at 20–30g to recover the beads. 9. Add 60 µL protein loading buffer to the beads. Incubate the samples at 90°C for 5 min to elute the bound proteins. Electrophorese 20 µL of each sample on a 12% SDS-PAGE gel, including 10 µg protein extracts as a positive control.

310

Cumming and Graham

10. Electroblot the gel onto nitrocellulose or PVDF membrane at 200 mA for 2 h or at 100 mA overnight at 4°C. Stir the buffer and use an ice block to prevent overheating. 11. Rinse the membrane briefly in PBS-T. 12. Block the membrane for at least 1 h on an orbital shaker at 4°C in PBS-T + 5% dried-milk powder. 13. Wash the membrane on an orbital shaker for 3 × 5 min with PBS-T. 14. Dilute the primary antibody in PBS-T + dried-milk powder (see Note 12). Prepare at least 1 mL of antibody solution per 10 cm2 of membrane. 15. Seal the membrane and antibody solution in a plastic bag, taking care to avoid leaks or air bubbles. Tape the bag to the platform of an orbital shaker. 16. Shake the membrane at high speed for 1 h at room temperature. 17. Remove the membrane from the bag and wash it for 6 × 5 min in PBS-T on an orbital shaker (see Note 13). 18. Dilute the secondary antibody as required in PBS-T + 5% dried-milk powder. Use an HRP-conjugated anti-mouse antibody, HRP-conjugated anti-rabbit antibody, or HRP-conjugated protein A, depending on the species in which the primary antibody was generated. Prepare at least 1 mL of secondary antibody solution per 10 cm2 of membrane. 19. Seal the membrane in a plastic bag, and incubate it on an orbital shaker for 1 h at room temperature. 20. Wash the membrane for 6 × 5 min in PBS-T. 21. Detect the signal with ECL reagent, blot the excess liquid from the membrane and seal it in plastic. Expose the membrane to X-ray film for 1 min. Repeat the exposure for a longer or shorter time if necessary (see Note 14).

4. Notes 1. Tritiated chloramphenicol is supplied in ethanol. Before use, take 1 µL and add to 2 mL scintillation fluid. Transfer the remainder to a 1.5-mL microcentrifuge tube, chill on dry ice for 5 min, and then evaporate the ethanol by drying under vacuum. Resuspend the tritiated chloramphenicol in 500 µL TE buffer. Working in a fume hood, add 500 µL xylene, vortex to mix thoroughly, and then centrifuge the sample for 2 min at full speed in a microcentrifuge. Remove and discard the upper layer; retain the lower aqueous layer. Extract once more with xylene, again retaining the lower layer. Add 500 µL petroleum ether, mix by vortexing, and then centrifuge for 2 min at full speed in a microcentrifuge. Retain the lower layer. Allow the excess ether to evaporate in the fume hood for 20 min, then add 1 µL of the sample to 2 mL scintillation fluid. Count the two samples on a liquid scintillation counter to estimate recovery, bearing in mind that the postpurification sample is approximately half the volume. Store the tritiated chloramphenicol at –20°C in small-volume aliquots to avoid repeated freezethawing. We have also used [14C]-chloramphenicol (NEN NEC408A) in these CAT assays. The counts obtained are lower, reducing the sensitivity of the assay. However, the reagent is supplied in aqueous solution and may be stored at 4°C, making it more convenient to use.


311

2. Solutions for these RNA–protein binding experiments should be maintained RNase free. Solutions (with the exception of Tris buffers) may be treated with 0.1% DEPC. In a fume hood, add two drops DEPC per 100 mL of solution and leave at room temperature overnight, then autoclave it. DEPC degrades to CO2 and H2O. RNase-free microcentrifuge tubes and pipet tips may be purchased, or may be prepared by soaking overnight in DEPC-treated water, and then autoclaved and baked dry. Solutions and plasticware for RNA experiments should always be kept separate from those used for other experiments. 3. It is important to be sure that your reporter assay results lie on the linear range for the assay, and that none of your reagents has become depleted. If all counts obtained seem very high and there is no clear difference between samples with and without putative NRE sequences, repeat the assay diluting the cell lysate. If all counts are very low, first confirm that the transfection has worked by assaying the samples for β-galactosidase. If the problem appears to lie with the CAT assay, first try preparing fresh chloramphenicol, since the stock solution may be unstable. If the assay still works very poorly, a new batch of butyryl coenzyme A may be required. 4. In order to make mutations by PCR, use conventional forward and reverse (outer) primers that will amplify the whole region of interest, including the NRE. For convenience, the fragment may be cloned into a plasmid such as pBluescript (Stratagene), providing convenient restriction sites and allowing outer T3 and T7 primers to be used. In order to introduce one or a small number of base changes into the element, design a pair of complementary inner primers containing a mismatch towards the center that will introduce the desired sequence (see Fig. 2). Alternatively, to make a deletion of 1 h at 80°C in a hybridization oven.

3.3.4. Probing the Southern Blot 1. Digest 2 µg of pOri16M with 10 U of PvuII and run on a 0.5X TBE, 1% agarose gel. 2. Cut the agarose containing the resultant 700-bp fragment out of the gel and purify the DNA using the QIAquick Gel Extraction Kit. 3. Using the Stratagene Prime-it II® random primer labeling kit according to the manufacturer’s instructions, label 25 ng of the DNA fragment by incubating at 37°C for 10 min in a reaction containing random 9mer primers, 5U of Exo Klenow DNA polymerase, 200 µM dATP, dGTP, and dTTP, and 5 µL of Redivue α 32P dCTP (50 µCi). 4. Purify the probe using a NICK Column according to manufacturer’s instructions. The purified probe is eluted with 400 µL of TE buffer and stored at –20°C in a lead-lined pot. 5. Incubate the nylon membrane prepared under Subheading 3.3.3. with 15 mL of QuickHyb hybridization solution in a conical hybridization tube. Revolve in a hybridization oven for 1 h at 68°C. 6. Incubate the probe for 5 min in a boiling-water bath and mix 100 µL with 1 mL of warm Quickhyb solution.

Quantitation of HPV DNA Replication

357

7. Add the probe to the hybridization tube and incubate at 68°C for 1 h. The probe hybridizes with pOri16M and all pCMV-derived vectors. 8. Wash the membrane twice for 15 min at room temperature with 2X SSC/0.1% (w/v) SDS solution. 9. Wash the membrane once for 40 min at 60°C with a 0.1X SSC/0.1% (w/v) SDS solution. 10. Seal the membrane in Saran wrap and expose to a phosphor screen overnight.

3.3.5. Analysis and Quantitation of the Southern Blot 1. Develop the image stored on the phosphor screen using a Storm™ 840 image system and visualize using ImageQuant™ software. 2. Measure the strength of replication by measuring the intensity of the bands representing the replicated DNA molecules.

3.4. Using Real-Time PCR Analysis to Quantitate E1- and E2-Mediated DNA Replication The use of Southern blotting for the analysis of E1- and E2-mediated DNA replication is far from ideal. This technique is hazardous, due to radioactivity; time consuming, as it takes 2 wk from initial setting up of cells until result; semi-quantitative, as a result of differential transfer of DNA onto membrane from the gel and from differential hybridization of the probe to the membrane; and sample limiting, as only so many samples can be run down an agarose gel at one time. To overcome all of these problems, we have developed a real-time PCR assay for the detection of E1- and E2-mediated DNA replication. This technique is an adaptation of the Southern blotting technique; the cell transfection and the harvesting of the low-molecular-weight DNA is carried out exactly as described under Subheadings 3.1.1., 3.1.2., and 3.2.1. See Note 5 for a discussion of when to use either of the techniques and Notes 6–8 for a description of some of the problems that can arise and their resolutions. Figure 3 represents the results of a typical experiment carried out in C33a cells and analyzed using this technique.

3.4.1. Preparation of the Harvested DNA for Real-Time PCR Analysis 1. Digest 25 µL of the DNA harvested from the cell and prepared as described under Subheading 3.2.1. overnight with Dpn1 in a final volume of 50 µL. 2. The following morning, treat the Dpn1-digested sample with 100 U of exonuclease III for 30 min. 3. Incubate the reaction for 30 min at 70°C to de-activate the exonuclease III.

3.4.2. Real-Time PCR Analysis of Harvested and Enzyme-Treated DNA 1. Incubate 30 µL of the sample prepared under Subheading 3.4.1. in a solution containing 5.5 mM MgCl2; 200 µM dATP, dCTP, and dGTP; 400 µM dUTP;

358

Morgan and Taylor

Fig. 3. An example of the results obtained using the real-time polymerase chain reaction (PCR) technique to monitor DNA replication. C33a cells were transfected with pOri16M, pCMV-E1, and pCMV-E2 as indicated (see Table 1 also). 72 h later, low-molecular-weight DNA was harvested and digested with Dpn1 and exonuclease III. The replicated pOri16M was detected in these samples using real-time PCR, and the values obtained are graphed as pg on an exponential scale.

900 nM of each primer; 100 nM probe, 1 U Amplitaq®, in 1X Geneamp® buffer at a final volume of 150 µL (see Note 9 for details of the primers and probe). 2. Aliquot 50 µL of this into 3 × well in a 96-well real-time plate. 3. Measure the quantity of pOri16M DNA detectable in the sample by assaying the real-time PCR plate on an ABI Prism 7700 using the universal real-time PCR conditions (95°C 10 min hot-start, 95°C for 30 s, and 60°C for 1 min, in 40 cycles).

3.4.3. Analysis of the Real-Time PCR Data 1. Analyze the real-time PCR data using the Sequence Detector 1.7 software. This calculates the standard curve (from 12 steps of 100 pg to 10–5 pg of pOri16M) and the quantity of DNA for each individual sample well.


359

2. Run each digested sample (see Subheading 3.4.2.) in triplicate. For each sample, calculate the final quantity of DNA present by taking the average of the two least divergent samples. This undigested pOri16M represents the replicated molecules.

4. Notes 1. If there is nothing detectable on the Southern blot, check that your technique is working by simply adding defined amounts of linearized pOri16M on a gel and Southern blotting. It should be easy to detect down to 4 pg of DNA using this technique; if you are not, then this problem must be solved before proceeding further. The problem could be probe preparation, gel preparation, transfer to membrane, or hybridization of probe to membrane. 2. If there is detectable input DNA on the Southern blot but there are no replicated molecules, then a range of E2 concentrations should be tried when first attempting the DNA replication assay. Too little E2 co-transfected will result in no replication; too much and there is squelching that again results in no, or poor, DNA replication. A typical window of concentrations to try would be 1 ng, 10 ng, 100 ng, and 1000 ng. The pOri16M can also be limiting, and it is also advisable to test a variety of pOri16M concentrations. 3. If there is detectable input DNA on the Southern blot but there are no replicated molecules, and the assay has been tried with a variety of E2 and pOri16M concentrations, remember that this is a replication assay, and therefore the cells must be healthy and dividing in order for good replication to occur. It is necessary to test any new cell line at a number of cell densities in this assay. If too dense or sparse, the cells will not replicate well and will also be susceptible to increased toxicity following the calcium phosphate precipitation technique. 4. A smear or unexpected bands down one or more of the lanes usually indicates a partial digest with either the Xmn1 or the Dpn1; test that these enzymes are still working. Occasionally this can be due to the actual sample itself; care must be taken when doing the DNA preparation to follow the protocol detailed above, in particular to ensure that the phenol/chloroform extraction proceeded well and that the aqueous/phenol interface is not disrupted during removal of the aqueous phase. Additionally, incomplete digestion may occur when too much DNA is digested in too small a volume. A simple remedy for this is to digest the DNA in a larger total volume and then ethanol precipitate the digested DNA to concentrate it for loading onto an agarose gel. Incomplete DNA digestion can be identified by two methods: a. When the agarose gel is UV-illuminated, samples that have been sufficiently digested will appear to have a faint maximal intensity just above 12 kb and the DNA will smear down the gel (see Fig. 4). In samples where the digest has been poor, there will be a strong band of cellular DNA just above 12 kb with a lesser smear down the gel (see lanes 13, 15, 16, and 18 in Fig. 4). These observations highlight the efficiency of Xmn1 digestion of cellular DNA; however, they closely reflect the efficiency of Dpn1 digestion also.

360

Morgan and Taylor

Fig. 4. Image of an agarose gel prepared for Southern blot. Hirt-extracted DNA was treated with Xmn1 and Dpn1, and the digested DNA was separated on an agarose gel. The DNA was visualized by ethidium bromide/ultraviolet illumination. Loaded on the left of the gel is a 1-kb DNA marker ladder, and the sizes of selected bands are shown in kb. Lanes 1–18 contain DNA digested with Dpn1 and Xmn1. The restriction-enzyme digest has worked efficiently in all lanes except 13, 15, 16, and 18. See Note 4 for an explanation of this potential problem as well as possible solutions. b. A probe that can hybridize to both pOri16M- and pCMV-derived vectors ensures that the replicated pOri16M observed on a Southern blot of Xmn1/ Dpn1-digested DNA is due to bona fide replication and not due to a failure of the Dpn1 digest. On blots where there is an incomplete Dpn1 digestion, there will be digested bands at the bottom of the gel, suggesting that the digest has worked; however, at approx 5 kb there will be bands representing pCMV-E1, pCMV-E2, or pCMV. Only when there are no CMV-derived bands on the upper portion of gel can you be confident that the Dpn1 digest has worked (see Fig. 2). 5. There are a few factors that should be considered when deciding to use either the Southern blotting or real-time PCR technique. There are clear advantages to using real-time PCR, as it is quantitative, non-toxic, and much easier and faster to process than Southern blotting. However, there are occasions where Southern blotting may be the only method available. In the real-time PCR procedure, a high background can be detected when elevated levels of pOri16M are used (100 ng to 1 µg), and therefore cell lines in which E1 and


6.

7.

8.

9.

361

E2 replicate DNA poorly and require elevated levels of pOri16M may not be suitable for the real-time PCR protocol. See ref. 7 for an in-depth description of the real-time PCR protocol and an example of standardizing a cell line for use with this technique. In the real-time PCR technique, if there is no detectable increase in signal in samples where DNA replication is predicted, then DNA replication may not have occurred, due to the reasons discussed in Notes 2 and 3. It is advisable to carry out a Southern blot to check whether this is the case. There are three possible reasons for too high a signal following PCR in the no replication control: (1) the Dpn1 and/or ExoIII digest did not work properly, (2) there was too much pOri16M used in the transfection, and (3) there could be contamination. Check that the enzymes are working by treating DNA with the enzymes and visualizing the DNA on a gel to confirm that it gives the expected pattern. Always carry out a titration with the pOri16M in the transfection to determine the levels that do not give too high a background signal. Finally, check that the reagents used in the protocol are not contaminated, by carrying out a reaction in the absence of added DNA. To enhance the quantitation of the real-time PCR technique it is possible to monitor the amounts of input plasmid present in the DNA sample harvested from the transfected cells. This serves to correct for transfection efficiency and efficiency of DNA harvest. To do this, the harvested DNA is treated with the restriction enzyme Mbo1; this enzyme recognizes the same sequence as Dpn1 with the crucial difference that it digests only nonmethylated DNA and therefore digests the freshly replicated DNA while maintaining the input DNA. The amount of this DNA can then be monitored by Southern blotting or the real-time PCR technique. Results obtained with the Dpn1 treatment can then be expressed relative to the levels of Mbo1 DNA detected; this controls for transfection and DNA-harvesting efficiency. An example of this approach is detailed in Taylor and Morgan (7). The design of primers and probes was carried out using Primer Express software (Applied Biosystems). The primer set chosen amplifies a 99-bp region of the HPV16 Ori cloned into pOri16M and has the Dpn1 site at the 3' end of the probe-binding site. This Dpn1 site was introduced using PCR; see ref. 6 for details.

Acknowledgments E. R. T. was supported by a studentship from the Biotechnology and Biological Sciences Research Council (BBSRC). The Royal Society also supported this work. References 1. LaPorta, R. F. and Taichman, L. B. (1982) Human papilloma viral DNA replicates as a stable episome in cultured epidermal keratinocytes. Proc. Natl. Acad. Sci. USA 79, 3393–3397.

362

Morgan and Taylor

2. Desaintes, C. and Demeret, C. (1996) Control of papillomavirus DNA replication and transcription. Semin. Cancer Biol. 7, 339–347. 3. Bouvard, V., Storey, A., Pim, D., and Banks, L. (1994) Characterization of the human papillomavirus E2 protein: evidence of trans-activation and trans-repression in cervical keratinocytes. EMBO J. 13, 5451–5459. 4. Steger, G. and Corbach, S. (1997) Dose-dependent regulation of the early promoter of human papillomavirus type 18 by the viral E2 protein. J Virol. 71, 50–58. 5. Vance, K. W., Campo, M. S., and Morgan, I. M. (1999) An enhanced epithelial response of a papillomavirus promoter to transcriptional activators. J. Biol. Chem. 274, 27,839–27,844. 6. Del Vecchio, A. M., Romanczuk, H., Howley, P. M., and Baker, C. C. (1992) Transient replication of human papillomavirus DNAs. J. Virol. 66, 5949–5958. 7. Taylor, E. R. and Morgan, I. M. (2003) A novel technique with enhanced detection and quantitation of HPV-16 E1- and E2-mediated DNA replication. Virology 315, 103–109. 8. Boner, W., Taylor, E. R., Tsirimonaki, E., Yamane, K., Campo, M. S., and Morgan, I. M. (2002) A functional interaction between the human papillomavirus 16 transcription/replication factor E2 and the DNA damage response protein TopBP1. J. Biol. Chem. 277, 22,297–22,303.

Analysis of E7/Rb Associations

363

26 Analysis of E7/Rb Associations Sandra Caldeira, Wen Dong, and Massimo Tommasino Summary The product of the early gene E7 is one of the major transforming proteins of human papillomaviruses (HPVs). It exerts its activity by associating with and altering the biological functions of several cellular proteins involved in the control of fundamental events, such as cell proliferation and apoptosis. The best-characterized activity of E7 from HPV type 16, the most frequently detected type in cervical cancer, is its ability to bind and induce degradation of the tumor-suppressor retinoblastoma protein (pRb) via the ubiquitin pathway. pRb plays a key role in cell-cycle control by negatively regulating, via direct association, the activity of several transcription factors, including members of the E2F family. The neutralization of pRb functions mediated by E7 results in constitutive activation of the transcription factors, with consequent loss of cell-cycle control. Several studies have shown that the oncogenic potential of a specific HPV type is dependent on the efficiency of E7 in targeting pRb. In this chapter, we describe two methods to measure the efficiency of the E7 proteins from different HPV types in neutralizing the pRb functions. The first one, the plate-binding assay, allows the determination of the pRb binding affinity of E7 proteins, while the second one permits the analysis of their impact on the pRb pathway in intact cells.

1. Introduction E7 is one of the major transforming proteins of human papillomaviruses (HPV) (1). Although more than 100 HPV types have been identified, only a few E7 proteins from the high-risk mucosal HPV types have been extensively characterized. In particular, the majority of studies have focused on E7 from HPV 16, since this is the most commonly detected HPV type, not only in invasive cervical carcinoma (ICC), but also in its precursor lesions, cervical intraepithelial neoplasia (CIN) (2,3). HPV-16 E7 is a small, acidic phosphoprotein that is functionally related to a gene product of another DNA tumor virus, the adenovirus (Ad) E1A protein (4,5). On the basis of the similarity in primary structure between the two viral From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

363

364

Caldeira, Dong, and Tommasino

Fig. 1. Schematic diagram of human papillomavirus (HPV)-16 E7 protein. The position of the three conserved regions (CR) is indicated by the numbers at the bottom of the figure. CR2 contains the LXCXE domain (amino acids 22–26), which mediates the association with the pocket proteins, and two serines at positions 31 and 32, which are phosphorylated by casein kinase II. CR3 contains two CXXC motifs that are involved in zinc binding and in protein stabilization.

proteins, they can be divided into three domains: conserved regions (CR) 1, 2, and 3 (Fig. 1). Mutational analysis of HPV-16 E7 has shown that integrity of all three CRs is essential for its biological functions. The transforming properties of HPV-16 E7 have been characterized in different cell types. Immortalized rodent fibroblasts, such as NIH 3T3, are transformed by E7 and acquire the ability to grow in serum-deprived medium and in soft agar (6). In addition, HPV-16 E7 alone, or in cooperation with E6 with higher efficiency, is able to immortalize primary human keratinocytes, which are the natural host cells of the virus (6). These properties of E7 are due mainly to its ability to associate with and inactivate the members of the pocket protein family, which is comprised of the product of the tumor-suppressor retinoblastoma gene (pRb) and two related proteins, p107 and p130 (1). The pocket proteins are key negative cell-cycle regulators (7), and their interaction with E7 leads to loss of cell-cycle control, favoring the exit of quiescent cells from G0 and entry into the S phase. Several studies have shown that the determination of the efficiency of E7 in inactivating pRb is a valid approach to assessing the oncogenic potential in


365

vivo of a specific HPV type. Comparative analyses between the different HPV types have revealed that the E7 proteins can be divided into three groups, according to their ability to target pRb: (a) E7 proteins that do not bind pRb, or bind with low affinity (8); (b) E7 proteins that bind pRb with high affinity (9,10); and (c) E7 proteins that bind pRb with high affinity and promote its rapid degradation via the proteasome pathway (11,12). The data available to date indicate that the induction of pRb degradation is a feature of the E7 proteins from the oncogenic HPV types. In this chapter, we describe assays for the determination of the efficiency of E7 in inactivating pRb. 2. Materials 2.1. Plate-Binding Assay 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26.

pET-21a+ bacterial expression vector (Novagen/Calbiochem). pGEX2T bacterial expression vector (Amersham/Pharmacia). Protease-deficient Escherichia coli strain BL21 (Life Technologies). Standard molecular-biology reagents (any source). Bacterial transformation kit (any source). Luria-Bertani (LB) medium. Ampicillin. Isopropyl-β-D-thio-galactopyranoside (IPTG). Bind/wash buffer: 4.3 mM Na2HPO4, 1.5 mM KH2PO4, 2.7 mM KCl, 140 mM NaCl, 0.1% Tween-20, 0.002% NaN3 (pH 7.3). Sonicator. Sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) equipment. T7-tag-specific monoclonal antibody (Novagen/Calbiochem). T7-tag purification kit (Novagen/Calbiochem). Elution buffer: 0.1 M citric acid (pH 2.2). Anti-mouse immunoglobulin (Ig)G antibody conjugated with horseradish peroxidase (HRP) (Promega). Bovine serum albumin (BSA). 30 mM Tris-citric acid (pH 6.0). Coomassie Brilliant Blue R-250 (Pharmacia). Carbonate buffer (pH 9.6): 1 volume of Na 2CO 3 (0.05 M) and 4 volumes NaHCO3 (0.05 M). Glutathione-casein (13). Ninety-six microwell PolySorp™ plates (Nunc). Blocking buffer: 0.2% casein, 0.05% Tween-20 in phosphate-buffered saline (PBS). Tetramethylbenzidine (Sigma). Hydrogen peroxide (30%). 1 M sulfuric acid. Enzyme-linked immunosorbent assay (ELISA) reader.

366


2.2. Determination of E7-Induced Degradation of pRb in NIH 3T3 Cells 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15.

16. 17. 18. 19. 20. 21. 22. 23.

Retroviral vectors, e.g., pLXSN (Clontech) and pBabe (14). Bacteria strain STBL2 (Life Technologies). Bacterial transformation kit (any source). LB medium. Ampicillin. Standard molecular-biology reagents (any source). Rodent fibroblasts (NIH3T3, ATCC No. CRL 1658). Packaging cell line Bosc 23 (ATCC No. CRL 11270). Chloroquine (Sigma). 2X HBS: 50 mM HEPES (pH 7.12), 10 mM KCl, 12 mM dextrose, 280 mM NaCl, 1.5 mM Na3PO4). CaCl2. Fetal calf serum (FCS) (any source; test before use). Dulbecco’s modified Eagle’s medium (DMEM). Polybrene (Sigma). IP buffer: 20 mM Tris-HCl (pH 8.0), 200 mM NaCl, 0.5% Nonidet P40, 1 mM ethylenediaminetetracetic acid (EDTA), 10 mM NaF, 0.1 M Na3VO4, 1 mM phenylmethylsulfonyl fluoride, 1 µg/mL leupeptin, 10 µg/mL soybean trypsin inhibitor, 10 µg/mL tosyl phenylalanine chloromethyl ketone, 1 µg/mL aprotinin, 10 µg/mL tosyl lysine chloromethyl ketone. Lowry-based assay to determine protein concentration (any source). SDS-PAGE equipment. Polyvinylidene difluoride membrane (NEN Life Sciences). Trans-Blot semidry electrophoretic transfer cell (Bio-Rad). Anti-hemagglutinin (HA) epitope antibody (MMS-101R, Babco). Anti-pRb (14001A, Pharmingen). Anti-β-tubulin (TUB2.1, Sigma). Anti-mouse IgG antibody conjugated with HRP (Promega).

3. Methods 3.1. In Vitro Assays to Determine E7-pRb Interactions Several methods have been developed to determine the efficiency of E7 proteins from different HPV types in binding pRb—e.g., glutathione S-transferase (GST)-pull-down assay, yeast two-hybrid assay, and plate-binding assay. The GST-pull-down assay is widely used in the HPV field and involves the use of bacterial recombinant E7 proteins. The E7 genes are fused in-frame to the carboxy-terminal sequence of the Schistosoma japonicum GST and expressed in bacteria. The recombinant proteins are immobilized on sepharose glutathione beads and incubated with cellular extracts. After extensive washing of the sepharose beads, the amount of pRb associated with E7 is determined by immunoblotting. A schematic representation of this assay is shown in Fig. 2A.


367

The yeast two-hybrid assay takes advantage of the fact that several eukaryotic transcriptional activators have two physically separable and functionally independent domains—the DNA-binding domain (DNA-BD) and the transcriptional activation domain (TAD). Fusion of these domains to two interacting proteins will result in full reconstitution of the transcriptional activator that is able to bind a specific promoter and activate the transcription of a reporter gene, such as lacZ (Fig. 2B). To characterize the E7-pRb interaction by a yeast two-hybrid assay, the E7 protein is fused to the transcriptional activation domain, while pRb is fused to the DNA-binding domain (for an example, see ref. 15). It is important to keep in mind that HPV-16 E7 has an intrinsic transactivation activity, and it is essential to fuse E7 to the TAD, since its fusion to the DNA-BD will result in activation of the reporter gene independently of the interaction with pRb. Like the GST pull-down assay, the plate-binding assay uses recombinant bacterial proteins. The plate is first saturated with casein, to which glutathione has previously been covalently linked (13). As a second step, a bacterial extract containing GST-pRb fusion protein is added to the plate. The bacterial proteins are removed by extensive washing, while the GST-pRb fusion protein remains immobilized on the plate. Bacterial recombinant E7 protein fused at the N-terminus to the T7 tag is finally added, and after washing, the amount of E7 associated with pRb is determined using a specific anti-T7-tag monoclonal antibody (all steps are summarized in Fig. 2C). In contrast to the two previous methods, the plate-binding assay offers the possibility to quantify the affinity for pRb of E7 proteins from different HPV types. For this reason, only this method is described in detail below.

3.1.1. Production of E7 proteins The E7 genes are cloned by standard techniques in-frame and downstream of the T7-tag sequence that is located in the multi-cloning site of the bacterial expression vector pET21a+. The fusion of E7 with the T7 tag is essential for purification of the recombinant protein (see Subheading 3.1.2.). In addition, the presence of a tag allows comparison of the pRb-binding affinity of different E7s, because the same monoclonal antibody can be used in the different assays. After completion of the cloning, perform the following steps: 1. Transform the protease-deficient strain BL21 with the different E7 constructs using standard techniques, e.g., electroporation or heat-shock protocol. 2. After transformation, plate the bacteria on LB medium containing 100 mg/L of ampicillin and grow at 37°C overnight. 3. Use a single colony to inoculate 20 mL of LB/ampicillin medium and grow overnight at 37°C.

368



369

4. Dilute 2 mL of overnight stationary phase culture in 200 mL of LB/ampicillin medium and grow at 37°C until the absorbance reaches 0.5–0.6 at 600 nm. 5. Cool the culture to a temperature of 30°C. 6. Initiate the expression of the E7 fusion protein by addition of IPTG to a final concentration of 1 mM. 7. After 4–5 h at 30°C, harvest the bacteria by centrifugation (5000g, 5 min) and freeze at –20°C or immediately process for purification of the recombinant proteins (see Note 1). 8. To purify the E7 proteins, resuspend the bacteria in 2 mL of bind/wash buffer and lyse them by sonication on ice (30 pulses of 2 s interrupted by 1-s intervals, microtip amplitude 30%). 9. After centrifugation of the bacterial lysate (16,000g, 5 min), purify the T7-E7 fusion proteins batch-wise using the T7-tag purification kit. Briefly, wash 200 µL of a 50% (v/v) agarose beads suspension with the T7 antibody twice with bind/wash buffer and add 1 mL of bacterial extract. After incubation for 30 min at 4°C in a rotator, collect the beads by centrifugation (see Note 2). Wash the beads five times with bind/wash buffer and elute the bound T7-tagged protein by adding 300 µL of elution buffer. Neutralize the buffer after elution by adding 45 µL of 1 M Tris base (pH 10.4). When larger amounts of purified protein are required, column purification can be performed (see Note 3). 10. Determine the concentration of the purified E7 fusion proteins spectrophotometrically at 280 nm using a calibration curve obtained with serial dilutions of BSA dissolved in 30 mM Tris-citric acid (pH 6.0). 11. Check the purity of the recombinant proteins by SDS-PAGE. One or two micrograms of purified recombinant proteins are applied onto a 15% SDS-polyacrylamide gel and visualized by Coomassie Brilliant Blue R-250 staining. We have successfully synthesized several T7-E7 fusion proteins from different HPV types. The results of a representative experiment are shown in Fig. 3, in which five different E7 proteins were purified.

Fig. 2. (opposite page) Schematic representation of three pRb-E7 association assays. (A) glutathione S-transferase (GST) pull-down assay. The GST-E7 fusion proteins are bound to glutathione-sepharose 4B beads. After washing, the sepharose beads are incubated with mammalian cellular extracts, and the levels of the pRb bound to GST-E7 are determined by Western blotting. (B) Yeast two-hybrid assay. The E7 protein is fused to the transcriptional activation domain (TAD), while pRb is fused to the DNAbinding domain (DNA-BD). Transactivation of the reporter gene (e.g., β-galactosidase) can occur only when the two proteins interact. (C) Plate-binding assay. GST-pRb is bound to a plastic surface previously coated with glutathione-casein. T7-E7 fusion proteins are added to each well, and the unbound fusion proteins are removed by extensive washing. Finally, the level of T7-E7 protein associated with pRb is quantified using a murine monoclonal T7-tag antibody and a secondary HRP-conjugated anti-mouse immunoglobulin antibody.

370


Fig. 3. Purification of bacterial recombinant T7-E7 proteins. T7-E7 recombinant proteins of HPV types 10, 16, 38, 48, and 60 were purified using anti-T7 tag antibody sepharose beads. Two micrograms of each recombinant protein were applied onto 15% sodium dodecyl sulfate-polyacrylamide gel and stained with Coomassie Blue. The positions of the molecular-mass markers in kDa are indicated on the left of the figure.

3.1.2. Production of Bacterial Lysate Containing GST-pRb Fusion Protein To generate the GST-pRb fusion protein, part of the pRb gene that encodes the E7-interacting domain (also termed the pocket domain) is cloned in the bacterial expression vector pGEX2T. It is not necessary to use the full-length pRb gene, since it yields much less fusion protein and does not increase the sensitivity of the assay. We have found that a fragment covering the coding region from amino acid 373 to amino acid 929 gives similar results to the fulllength protein in this assay for several E7 proteins. 1. Transform the protease-deficient strain of BL21 bacteria with the pGEX2TpRb construct and grow in LB/ampicillin medium as described under Subheading 3.1.1. 2. After the bacterial culture has reached an absorbance of approx 0.4 at 600 nm, initiate the induction of GST-pRb expression by adding IPTG to a final concentration of 0.1 mM. 3. Grow the bacteria for 3–4 h at 30°C and then lyse them by sonication as described in Subheading 3.1.1. (see Note 4).


371

3.1.3. Plate-Binding Assay Each assay should be performed in triplicate in separate plates. The assay comprises several steps, described in detail as follows: 1. Coat the 96-microwell plate with glutathione-casein by adding 100 µL of glutathione-casein solution to each well, and leave the plate overnight at 4°C (see Note 5). 2. Remove the glutathione-casein solution and add 200 µL of blocking buffer (for 1 h at 37°C or overnight at 4°C). 3. Remove the blocking buffer. 4. Wash the plate five times by immersion in a tank containing PBS/Tween-20 0.05%. Remove the washing solution by inverting and tapping the plate on paper towels. 5. Add bacterial extract containing GST-pRb fusion protein to each well. The amount of bacterial extract to use is dependent on the yield of GST-pRb protein, which can vary in different laboratories (see Note 6). 6. Incubate the plate at 4°C for 1 h. 7. Wash the plate as described in step 4. 8. Add 100 µL of purified bacterial recombinant T7-E7 fusion proteins (stock solution approx 500 ng/mL in blocking buffer). 9. Incubate the plate for 2 h at 4°C. 10. Wash the plate as described in step 4. 11. Add 100 µL T7 tag-specific monoclonal antibody (dilution 1/10,000 in blocking buffer) to determine the amount of T7-E7 protein associated with GST-pRb. 12. Incubate the plate at room temperature for 45 min. 13. Wash the plate as described in step 4. 14. Add the secondary anti-mouse IgG antibody conjugated with HRP (1/5000 dilution in blocking buffer). 15. Incubate the plate at room temperature for 45 min. 16. Wash the plate as described in step 4. 17. Start the colorimetric reaction by adding 100 µL of tetramethylbenzidine with 0.003% H2O2 as substrate. 18. Stop the enzymatic reaction by adding 50 µL of 1 M sulfuric acid to each well. 19. Determine the absorbance at 450 nm using an ELISA reader. For the controls, see the conditions described in Note 7.

3.1.4. Determination of KD KD is calculated using the equation KD = ([pRb] [E7])/[pRb-E7]. At [E750%], half of the pRb is free and half is bound to E7; thus, [pRb] = [pRb-E7] so that KD = [E750%]. The concentration of E7 giving 50% of the maximal specific optimal density (OD) [E750%] can be determined by fitting the experimental data (normalized OD values at different E7 concentrations) with a single rectangular hyperbola using the program SigmaPlot (SPSS Science, Chicago, IL).

372


Fig. 4. Human papillomavirus (HPV)-16 E7-pRb plate-binding assay. Increasing concentrations of HPV-16 E7 (from 0.7 pM to 0.8 µM) were added to the wells, in which a 40 nM solution of GST-pRb had been previously placed. The colorimetric reaction was developed for 5 min. The optimal density represents the normalized average of triplicate determinations. The line shows the fit of the experimental data with a single rectangular hyperbola (Sigma plot). The KD was determined using the formula described in the Methods section and corresponds to the amount of HPV-16 E7 (4.5 nM) sufficient to bind 50% of the pRb molecules immobilized on the well.

To facilitate the plotting of the data and to allow comparative analysis, normalize the OD values taking into consideration the maximum reachable OD value of a specific E7 as 100%. One representative experiment using HPV-16 E7 is shown in Fig. 4. In some cases, E7 protein can be 75% pure. However, even under these conditions, the calculated KD does not differ more than 25% from the real value.

3.2. Characterization of the Ability of E7 to Promote pRb Degradation in Immortalized Rodent Fibroblasts Immortalized rodent fibroblasts, such as NIH 3T3, represent an easy-tohandle and reliable model system to assay the efficiency of E7 in targeting pRb. Several studies have shown that HPV-16 E7 is able to bind and induce degradation of murine and human pRb with similar efficiency. Thus, E7-induced degradation can be easily monitored in NIH 3T3 cells.


373

Recombinant retroviral systems are widely used by several laboratories for expression of the E7 genes into target cells. The retroviruses are capable of delivering genes to target cells with high efficiency in a manner that allows integration of the introduced genetic elements and long-term, stable expression of the genes.

3.2.1. Generation of E7 Recombinant Retroviruses High-titer recombinant retroviruses carrying the desired genes are generated by transient transfection of the Bosc-23 ecotropic packaging cell line (capable of delivering genes to dividing murine or rat cells) (see Note 8). This packaging line is derived from 293-T cells (Ad5- and SV40-transformed, human, embryonic kidney 293 cell line) and contains the retroviral structural genes gag, pol, and env, as well as the origin of replication (ori) of Moloney murine leukemia virus (MMuLV). When transfected with the retroviral expression vector that provides the RNA packaging signal Y, plus transcription and processing elements, the cells are able to generate retroviral particles carrying the recombinant vector. The particles can then be used to infect the target cells and transmit the recombinant genes. However, they cannot replicate within the target cells, because these lack all the viral structural genes. As retroviral vectors, we have used pBabe (14) or pLXSN vectors (Clontech). To obtain the E7 genes, the viral DNA of interest is amplified by polymerase chain reaction (PCR) using primers that contain sequences that both flank the open reading frames (ORFs) and introduce restriction sites. Cloning is performed by standard recombinant DNA techniques (see Note 9). Few E7 antibodies are commercially available, and they do not recognize all E7 proteins. Therefore, to verify the expression of the different E7 genes, a specific epitope is added to the viral gene. We normally clone our E7 genes in-frame with the hemagglutinin (HA) tag at the N terminus. Our previous data showed that addition of HA tag to the E7 proteins does not alter the ability of the viral protein to promote pRb degradation.

3.2.2. Transient Transfection of the Packaging Cell Line Bosc 23 by Calcium Phosphate Precipitation The DNA used for transfection should be of high quality. We have obtained good results using the Maxi Prep kit from Qiagen. 1. Plate 2.5–4 × 106 cells on 10-cm dishes in 5–10 mL growth medium (DMEM + 10% FCS) and incubate for 18–24 h before transfection at 37°C, under 5% CO2 (see Note 10). 2. Transfect subconfluent cells (70%) with the empty or the recombinant retroviral vectors by calcium phosphate precipitation. Fifteen minutes before transfection, change the medium to fresh medium containing 2.5 µM chloroquine (see Note 11).

374

3. 4. 5. 6.

Caldeira, Dong, and Tommasino Prepare the following transfection cocktail: 10 µg of DNA + 62 µL 2 M CaCl2 + H2O to a final volume of 500 µL. Add this mix to 500 µL of 2X HBS (drop-wise with gentle agitation), (see Note 12). Spread the mix drop by drop across the whole plate. Be very gentle in agitating the plate (back and forth, left and right) and return it to the incubator for 8–10 h. Remove the chloroquine-containing medium and wash the cells twice with PBS to remove remaining chloroquine and precipitated DNA (see Note 13). Add 5 mL DMEM containing 10% FCS medium and incubate overnight. Change the medium 24 h after transfection. Collect the virus-containing medium at 48 h after transfection, when the production and release of virus is maximal. The medium is normally aspirated with a syringe and filtered through a 0.22-µm filter into a 5-mL Falcon tube and used to infect the target cells (see Note 14).

3.2.3. Infection of the Target Cells 1. Plate the target cells 24 h before infection in order to have approx 50% confluence the day after. The number of plates (10 cm diameter) to be prepared is (n + 1) for n infections. The extra plate is used as a negative control for the antibiotic selection after the infection (see also step 6 below). 2. Centrifuge the virus-containing medium for 5 min at 500g . If the viral supernatant was kept at –70°C, place it at room temperature and centrifuge it as soon as it is thawed. 3. Add polybrene to the viral supernatant to a final concentration of 4 µM (see Note 15). 4. When ready to infect, substitute the growth medium of the target cells by the medium containing the retrovirus prepared as described under Subheading 3.2.2. and return the plate to the incubator. 5. After 3 h, add 5 mL of fresh growth medium. 6. At 48 h after infection, select the cells with the appropriate antibiotics (see Note 16). At this time, the cells should be nearly confluent in the plate and must be split for selection. The split ratio depends on the growth rate of the cells and the efficiency of the infection, and has to be determined experimentally (see Note 17). Noninfected cells are exposed to the same concentration of antibiotics as a control to determine when selection is terminated. After selection, culturing is continued in antibiotic-containing medium.

3.2.4. pRb Degradation Assay in NIH 3T3 Cells For the pRb degradation assay, cells infected with HPV-16 E7 retroviruses can be used as a positive control, while we generally use empty vector retrovirus as a negative control. At the end of the selection, total cellular extracts are prepared to ascertain the expression of E7 protein and the levels of pRb. 1. Lysis of the cells should be performed at 75% confluence, so it may be necessary to split the cells during selection.


375

2. After removal of the medium, wash the cells twice in ice-cold PBS (see Note 18). 3. Carefully aspirate the PBS from the plate and lyse the cells in IP buffer (for 20 min at 4°C). We normally use 200–300 µL IP buffer per 10-cm dish. Centrifuge the extract at 13,000g for 15 min at 4°C. 4. Recover the supernatant and determine the protein concentration using a Lowrybased assay. 5. Because we fuse our E7 genes to the HA tag, we use an HA-tag monoclonal antibody to detect their expression. Western blotting is a standard molecularbiology technique and it will not be described here in detail. Briefly, we fractionate 100 µg of extract (see Note 19) by electrophoresis in a 15% SDSpolyacrylamide gel and transfer it to a polyvinylidene difluoride (PVDF) membrane (DuPont) using the Trans-Blot semidry electrophoretic transfer cell (see Note 20). 6. After the transfer, the membrane is washed twice in 5% milk powder/PBS and then incubated overnight at 4°C (or 2–3 h at room temperature) in the same buffer containing the anti–HA tag antibody (diluted 1/1000). The membrane is then processed following a standard immunoblotting protocol using a secondary antibody conjugated with HRP (see Note 21). 7. The ability of E7 to induce pRb degradation is evaluated by determining intracellular levels of pRb by immunoblotting using the same cellular extracts. Cellular extract (100 µg) is loaded onto an 8% SDS-polyacrylamide gel. 8. After protein transfer as described previously, immunoblotting is performed using an anti-pRb antibody (diluted 1/1000). To exclude the possibility that variations in pRb levels may be due to different loading onto the SDS-polyacrylamide gel of the different cellular extracts, the same membrane is incubated with an antibody of the product of a housekeeping gene, β-tubulin antibody (diluted 1/1000) (see Note 22). A representative experiment is shown in Fig. 5.

By comparing the amounts of pRb in extracts expressing the different E7 proteins as well as our positive (HPV-16 E7) and negative (empty vector) controls, it is possible to evaluate the ability of the E7s to promote pRb degradation. 4. Notes 1. We obtained similar yields for the majority of the E7 proteins, but in some cases low levels of recombinant proteins are obtained. Adjusting the time of induction of protein expression or the temperature of the bacterial culture may increase the yield. 2. Incubation at 4°C in this step reduces degradation, as no protease inhibitors were used. Alternatively, protease inhibitors can be used, as they do not influence the purification efficiency or subsequent assays. In this case, the incubation can be performed at room temperature to facilitate the antibody/antigen reaction. 3. If larger amounts of purified fusion protein are required, column purification can be performed. For this purpose, 2 mL of 50% (v/v) suspension of beads are

376


Fig. 5. Determination of pRb levels in NIH 3T3 expressing human papillomavirus (HPV)-1 or HPV-16 E7. One-hundred micrograms of protein extracts of cells expressing the different E7 proteins as indicated in the figure were applied on 8% (for pRb detection) or 15% (for E7 detection) polyacrylamide-sodium dodecyl sulfate gel, transferred onto polyvinylidene difluoride membrane, and incubated with an anti-HA tag, a pRb, or β-tubulin antibody. β-tubulin signal was used as loading control. applied on a chromatography column. Equilibrate the column with 10 volumes of bind/wash buffer and then apply 1 mL of bacterial extract. Wash the column with 10 volumes of bind/wash buffer and elute the bound proteins with five serial 1-mL volumes of elution buffer into five separate tubes containing 150 µL of 1 M Tris base (pH 10.4). The column can be recycled five times with minimal loss of binding activity. 4. It is important to check the expression of GST-pRb fusion protein before performing the plate-binding assay. The level of recombinant protein can be determined easily by Western blotting using an anti-GST antibody. The bacterial extract can be divided into aliquots, kept at –20°C, and used for several independent experiments. 5. Glutathione-casein can be synthesized as previously described (13) and dissolved in 50 mM carbonate buffer (pH 9.6) at a final concentration of 2 ng/µL. 6. For determination of the minimal saturation amount of GST-pRb, coat the microwell plate with glutathione-casein and, after blocking, add to the plate increasing concentrations of total protein extract of GST-pRb-expressing cells (from 1 µg/mL to 5 mg/mL). Wash off unbound bacterial proteins and incubate the plate at room temperature for 45 min with polyclonal rabbit GST antibody (1:2000 dilution in blocking buffer, Sigma). After washing (see Subheading 3.1.3., step 4), incubate the plate at room temperature for 45 min with secondary anti-rabbit IgG antibody HRP conjugate (1:5000 dilution in blocking buffer, Promega) and perform the colorimetric reaction. The minimal concentration of


7.

8.

9.

10.

11. 12.

13.

14. 15.

16. 17. 18.

19.

377

bacterial extract sufficient to saturate the well will correspond to the point at which the intensity of the colorimetric reaction has reached a plateau. To determine the specificity of the assay, two different controls are included, in which either GST-pRb or T7-E7 protein is replaced by 100 µg of BL21 protein extract. The protocol can be followed to transfect amphotropic packaging cell lines (e.g., Phoenix) and generate amphotropic retroviruses that are capable of delivering genes to dividing cells of most mammalian species, including human. It is recommended to use STBL2 bacteria (Life Technologies) for amplifying any retroviral plasmid, since these bacteria are engineered to stabilize repeats and retroviral sequences. The plating step is very critical for obtaining good transfection efficiency. Cells have to be very well homogenized so that they will not grow in clumps. In addition, Bosc 23 cells are quite fragile and detach very easily. Handle them gently. Chloroquine inhibits lysosomal DNAses, helping DNA, delivered by Ca2PO4, through the lysosomes. All solutions should be at room temperature at the time of transfection. 2X HBS is stored in aliquots at –20°C. Each aliquot is defrosted immediately before the transfection. The remaining 2X HBS solution can be kept at 4°C for at most 1 wk and used for an independent transfection. pH is extremely important. The main reason for failure of calcium phosphate transfections is poor HBS. Always test your batch in a pilot experiment. It is important to be very gentle during this operation to avoid detachment of the cells. Check the plate under the microscope. If sandy dots of precipitated DNA are still present in the plate, repeat the washing step. The medium containing the recombinant retrovirus can be frozen at –70°C and kept for many months. However, freezing and thawing reduces the viral titer. Polybrene neutralizes the negative charge present on the surface of the virus and the cell, thereby reducing electrostatic repulsion and increasing the infection efficiency. Puromycin or neomycin can be used at final concentrations of 2 µg/mL or 1000 µg/mL, respectively. As a reference, NIH 3T3 cells infected with HPV-16 E7 retroviruses can be easily split 1:5, when 90% confluent. It is extremely important to completely remove the culturing medium. Residual medium containing fetal calf serum will interfere with the determination of protein concentration of the cellular extracts. The protein concentration will be too high because of the fetal calf serum proteins in the medium. In some cases, the protein concentration is very low. In order to have a small enough volume for loading onto SDS-polyacrylamide gel, protein extracts can be concentrated by acetone precipitation. A protein extract containing the desired amount of proteins is mixed with nine volumes of ice-cold acetone and incubated for 20 min at –20°C. After centrifugation (10 min at maximum speed in a standard bench centrifuge), the pellet is directly resuspended in SDS loading buffer.

378


20. The transfer of the proteins onto the PVDF membrane is normally completed after 90 min at 130 mA. However, the conditions can vary from one laboratory to another. It is always best to check the efficiency of protein transfer by staining the membrane after the transfer with Ponceau red or using stained protein markers (e.g., Rainbow from Amersham). 21. We normally use as secondary antibodies anti-mouse or anti-rabbit IgG antibodies conjugated with HRP (Promega) diluted 1/5000. 22. pRb and β-tubulin have different molecular weights (approx 105 and 55 kDa, respectively). Therefore, immunoblotting for these two proteins can be performed in parallel, simply cutting the membrane into two parts.

Acknowledgments The authors thank all the members of the laboratory for their interest and cooperation in the work described here, and Dr John Cheney for English revision. References 1. Munger, K., Basile, J. R., Duensing, S., et al. (2001) Biological activities and molecular targets of the human papillomavirus E7 oncoprotein. Oncogene 20, 7888–7898. 2. Clifford, G. M., Smith, J. S., Plummer, M., Munoz, N., and Franceschi, S. (2003) Human papillomavirus types in invasive cervical cancer worldwide: a meta-analysis. Br. J. Cancer 88, 63–73. 3. Munoz, N., Bosch, F. X., de Sanjose, S., et al. (2003) Epidemiologic classification of human papillomavirus types associated with cervical cancer. N. Engl. J. Med. 348, 518–527. 4. Phelps, W. C., Yee, C. L., Münger, K., and Howley, P. M. (1988) The human papillomavirus type 16 E7 gene encodes transactivation and transformation functions similar to those of adenovirus E1A. Cell 53, 539–547. 5. Vousden, K. H. and Jat, P. S. (1989) Functional similarity between HPV16E7, SV40 large T and adenovirus E1a proteins. Oncogene 4, 153–158. 6. Mansur, C. P. and Androphy, E. J. (1993) Cellular transformation by papillomavirus oncoproteins. Biochim. Biophys. Acta 1155, 323–345. 7. Morris, E. J. and Dyson, N. J. (2001) Retinoblastoma protein partners. Adv. Cancer Res. 82, 1–54. 8. Munger, K., Werness, B. A., Dyson, N., Phelps, W. C., Harlow, E., and Howley, P. M. (1989) Complex formation of human papillomavirus E7 proteins with the retinoblastoma tumor suppressor gene product. EMBO J. 8, 4099–4105. 9. Giarre, M., Caldeira, S., Malanchi, I., Ciccolini, F., Leao, M. J., and Tommasino, M. (2001) Induction of pRb degradation by the human papillomavirus type 16 E7 protein is essential to efficiently overcome p16INK4a-imposed G1 cell cycle arrest. J. Virol. 75, 4705–4712.


379

10. Caldeira, S., Dong, W., Tomakidi, P., Paradiso, A., and Tommasino, M. (2002) Human papillomavirus type 32 does not display in vitro transforming properties. Virology 301, 157–164. 11. Boyer, S. N., Wazer, D. E., and Band, V. (1996) E7 protein of human papilloma virus-16 induces degradation of retinoblastoma protein through the ubiquitinproteasome pathway. Cancer Res. 56, 4620–4624. 12. Jones, D. L. and Munger, K. (1997) Analysis of the p53-mediated G1 growth arrest pathway in cells expressing the human papillomavirus type 16 E7 oncoprotein. J. Virol. 71, 2905–2912. 13. Sehr, P., Zumbach, K., and Pawlita, M. (2001) A generic capture ELISA for recombinant proteins fused to glutathione S-transferase: validation for HPVserology. J. Immunol. Meth. 253, 153–162. 14. Morgenstern, J. P. and Land, H. (1990) Advanced mammalian gene transfer: high titre retroviral vectors with multiple drug selection markers and a complementary helper-free packaging cell line. Nucleic Acids Res. 18, 3587–3596. 15. Ciccolini, F., Di Pasquale, G., Carlotti, F., Crawford, L., and Tommasino, M. (1994) Functional Studies of E7 proteins from different HPV types. Oncogene 9, 2342–2348.

Transformation Assays for HPV Oncoproteins

381

27 Transformation Assays for HPV Oncoproteins Paola Massimi and Lawrence Banks Summary A cornerstone of human papillomavirus (HPV) research was the demonstration that those HPV types associated with the development of cervical cancer encode two potent oncoproteins, while those HPV types associated with only benign lesions do not. Thus both HPV-16 E6 and E7 will transform established rodent cells and will efficiently cooperate with other activated oncogenes in the transformation of primary rodent cells. In addition, the virus also encodes for the E5 oncoprotein, which also possesses a weaker transforming activity in established rodent cells. In this chapter we describe how the transforming activities of the HPV oncoproteins can be assessed.

1. Introduction Carcinogenesis occurs through a multistep process, and multiple oncogenes are required for the full transformation of normal primary cells in vitro (1–3). Assays for assessing the transforming activity of the different human papillomavirus (HPV) oncoproteins fall into three categories: (a) transformation of established rodent cells; (b) transformation of primary rodent cells; (c) immortalization of primary human cells. A key feature in all cases is that only the high-risk HPV types score positive in these assays. The first indication that HPV 16 encoded a transforming activity came from studies of the viral early region expressed in retroviral constructs. In these studies, E7 was found to be the most potent oncogene (4–6), followed by E6, which can transform established rodent cells but with less efficiency (7,8). E5 was also found to possess transforming activity in 3T3 A31 cells (7), which was further stimulated by the addition of epidermal growth factor (EGF) (9,10). The use of primary rodent cells represents a more relevant assay with respect to the transforming activity of HPV in vivo, since the target cells are primarily epithelial in origin. Target cells can be primary baby rat kidney cells (BRK) or primary mouse kidney From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

381

382

Massimi and Banks

cells (BMK). The presence of a cooperating activated oncogene (typically EJ-ras) is required in these assays (11,12). In BRK cells, the principal transforming activity is encoded by the E7 oncoprotein in high-risk HPVs (13–16), while in BMK cells, E6 has a transforming activity close to or equivalent to that of E7 (17,18). Perhaps the most relevant system with respect to the activity of the HPV oncogenes in human tumors is the immortalization of primary human keratinocytes (from genital tract or oral mucosa). In these cells, E7 alone can induce hyper-proliferation, although they eventually undergo senescence (19–26). However, E6 and E7 from high-risk but not from low-risk HPV types efficiently induce immortalization of these cells and do not require the presence of another activated oncogene. Moreover, E7 is able to induce immortalization of human keratinocytes in the absence of E6, albeit at low frequency (27,28), and E5 has been demonstrated to stimulate mitogenesis enhanced by addition of EGF in primary human cells (29,30). In this chapter we describe the design and implementation of assays to measure the transforming activities of HPV E5, E6, and E7 in established rodent cells, as well as the oncogene cooperating activities of HPV E6 and E7 in primary rodent cells. A key element in these assays as well as those described in Chapter 29 is the quality of the DNA used to express the viral oncoprotein; therefore, we also include in this chapter a description of the method we use to prepare our plasmid DNA. 2. Materials 2.1. Cells 1. NIH 3T3, 3T3 A31 (established mouse fibroblasts) (American Type Culture Collection). 2. Baby mouse kidney primary cells derived from 9-d-old mice BALB/c (BMK) (Harlan-Italy). 3. Baby rat kidney primary cells derived from 9-d-old rats Wistar Hannover (BRK) (Harlan-Italy).

2.2. Cell Culture 1. Dulbecco’s modified Eagle’s medium (DMEM). 2. Supplemented DMEM: DMEM with 10% FBS, 1% glutamine, 200 U/mL penicillin, and 100 µg/mL streptomycin. 3. Sterile phosphate-buffered saline (PBS): 137 mM NaCl, 2.7 mM KCl, 1.5 mM KH2PO4, 8.1 mM Na2HPO4 (pH 7.4). 4. Versene buffer (1X): 1 mM EDTA, 0.17 M NaCl, 3 mM KCl, 10 mM Na2HPO4, 1.6 mM KH2PO4 (pH 7.2). 5. Trypsin/ethylenediamine tetraacetic acid (EDTA) 0.05% in 10X Versene. 6. 100-mm Tissue-culture dishes. 7. Inverted microscope.


383

2.3. DNA Preparation 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

Luria broth: 10 g Bacto-tryptone, 5 g Bacto-yeast extract, 10 g NaCl per 1 L. Ampicillin (50 mg/mL). Chloramphenicol (34 mg/mL in ethanol). Solution 1: 50 mM glucose, 25 mM Tris-HCl (pH 8), 10 mM EDTA. Solution 2: 0.2 N NaOH, 1% sodium dodecyl sulfate (SDS). Solution 3: 3 M potassium acetate (pH 4.8), 5 M glacial acetic acid. 2-propanol. Cesium chloride. Ethidium bromide (10 mg/mL). TE: 10 mM Tris-HCl (pH 8), 1 mM EDTA. 2.5-mL Syringes. 10-mL Syringes. 18-Gage needles. Sylanized corex glass tubes. Beckman Quick-Seal tubes. Heat-sealing instrument (Beckman). Beckman centrifuge (Ti70 rotor). Sorvall centrifuge (GS3, SS34 rotors). Spectrophotometer.

2.4. Transfection of Cells 1. TE: 10 mM Tris (pH 8.0), 1 mM EDTA. 2. 2X HEPES-buffered saline (HBS): 2.8 M NaCl, 250 mM HEPES (pH 7), 150 mM Na2HPO4 (pH 7.12) (see Note 1). 3. 2.5 M CaCl2. 4. 1X Versene. 5. Glycerol. 6. Expression plasmids (see Note 2).

2.5. Cell Colony Staining 1. Sterile PBS 2. Paraformaldehyde (10% in PBS). 3. Giemsa Blue stain (10% in PBS).

2.6. Establishing Primary Cell Cultures 1. Sterile PBS. 2. Serum free supplemented DMEM: DMEM with 1% glutamine, 200 U/mL Penicillin and 100 µg/mL Streptomycin. 3. Trypsin/EDTA 0.25% in Versene 10X. 4. Fetal bovine serum (FBS). 5. Sterile forceps and scissors. 6. 60-mm Tissue-culture dishes. 7. 100-mm Tissue-culture dishes.

384

Massimi and Banks

8. 50-mL Falcon tubes. 9. Geneticin (G418) (Sigma).

2.7. Picking Colonies 1. 2. 3. 4. 5. 6.

Sterile PBS. Trypsin/EDTA 0.05% in Versene 10X. Plastic cloning rings. Vacuum grease. 24-, 12-, and 6-well tissue-culture plates. 75-cm2 Tissue-culture flasks

2.8. Western Blots 1. Cell lysis buffer: 250 mM NaCl, 0.15% NP40, 50 mM HEPES (pH 7), 1% aprotinin, 100 µM TLCK, 200 µM TPCK, 1 mM phenylmethylsulfonyl fluoride (PMSF). 2. 2X SDS gel-loading buffer: 100 mM Tris-HCl (pH 6.8), 200 mM dithiothreitol (DTT), 4% SDS, 0.2% Bromophenol Blue, 20% glycerol. 3. Electrophoresis buffer: 25 mM Tris, 250 mM glycine—electrophoresis grade (pH 8.3), 0.1% SDS. 4. Transfer buffer: 39 mM glycine, 48 mM Tris base, 0.037% SDS, 20% methanol. 5. Nitrocellulose 0.22 µm (Schleier & Schuell). 6. Antibodies: e.g., anti-HPV E7 Ab (Santa Cruz); rabbit anti-mouse biotin conjugated (DAKO); avidin-peroxidase conjugated (DAKO). 7. Blocking solution: 10% powdered milk in PBS. 8. Antibody diluting solution: 10% powdered milk in PBS/0.5% Tween. 9. Washing buffer: 0.5% Tween in PBS. 10. ECL Kit (Amersham).

2.9. Assessing Anchorage-Independent Growth in Soft Agar 1. 2. 3. 4. 5.

60-mm Tissue-culture dishes. 5-mL Bijoux tubes. Difco Noble Agar. FBS. p-iodonitrotetrazolium violet.

3. Methods 3.1. Large-Scale Preparation of Plasmid DNA For any transformation assays as well as in vitro translation, the quality of the DNA used is critical. We suggest plasmid preparation using the following cesium chloride/ethidium bromide gradient protocol (31). 1. Plasmids are amplified by growing bacteria overnight at 37°C in 400 mL of Luria broth supplemented with the appropriate antibiotic (e.g., 50 µg/mL of ampicillin) with vigorous shaking in a rotary shaker (see Note 3).


385

2. The next day, the bacteria are pelletted at 12,500g in a GS3 Sorvall rotor using a polyallomer tube (or its equivalent) for 10 min. 3. The pellet is lysed in 18 mL of alkaline solution 1 (32) and left to stand at room temperature for 5 min. 4. Then 40 mL of freshly made solution 2 are added and the suspension mixed by gently inverting the tube several times. After an incubation on ice for 10 min, 20 mL of ice-cold solution 3 are added and the tube is again inverted several times until the liquid phases are no longer distinguishable and a flocculent white precipitate (comprising chromosomal DNA, high-molecular-weight RNA, and membrane complexes) is formed. 5. After 30 min on ice, the suspension is centrifuged for 10 min at 20,000g and the supernatant is then filtered through three layers of cheesecloth into a fresh tube containing 0.6 volume of 2-propanol. 6. The nucleic acids are recovered by centrifugation at 20,000g for 15 min, the supernatant is carefully decanted, and the bottle is left open to allow the pellet to dry. The DNA pellet is then dissolved in 11.5 mL of TE (pH 8). For every 1 mL of TE suspension, then add 1 g of solid CsCl. The solution is mixed gently to allow the salt to dissolve, and then 0.6 mL of ethidium bromide (10 mg/mL in water) are added for every 10 mL of the DNA/CsCl solution. Everything is centrifuged for 15 min at 20,000g in a Sorvall to eliminate the complexes formed between the ethidium bromide and the residual bacterial proteins. 7. Using a 10-mL syringe fitted with a large (18-gage) needle, the clear red solution is transferred into a Beckman Quick-Seal, or equivalent tube suitable for centrifugation, in a Beckman vertical Ti70 rotor. The tube is sealed using a heatsealing instrument and is centrifuged for 24 h at 150,000g at 24°C in vacuum. At the end of the centrifugation, two bands of DNA located in the center of the gradient are visible in ordinary light: the upper band, which should usually contain less material, consists of linear bacterial (chromosomal) DNA and nicked circular plasmid DNA; the lower band consists of closed circular plasmid DNA. The pellet at the bottom consists of ethidium bromide/RNA complexes (Fig. 1). The circular plasmid is collected in a fresh tube with an 18-gauge needle on a 2.5-mL syringe, after having inserted another needle in the top of the tube to allow air to enter. The ethidium bromide is removed through several extractions with a TE/2-propanol/Cs/Cl saturated solution, until the pink color disappears from both the aqueous and the organic phases. 8. DNA is then precipitated in a sylanized corex glass tube with 2 volumes of TE and 2.5 volumes of absolute ethanol. After an incubation of 3 h at –20°C, the tube is centrifuged at 25,000g in a SS34 Sorvall rotor for 20 min. The DNA pellet is dried, dissolved in sterile TE, and the final concentration measured by OD260 in a spectrophotometer.

3.2. Transfection of Established Rodent Cells In the context of HPV, the most potent viral oncoprotein in established rodent cells is E7. Together with plasmid DNA encoding HPV-16 E7, which

386

Massimi and Banks

Fig. 1. Scheme of plasmid DNA extraction. The upper band in the gradient contains linear bacterial DNA and nicked circular plasmid DNA; the lower contains closed circular plasmid DNA. The band on the top contains proteins, and the pellet on the bottom is made by ethidium bromide and RNA. The plasmid DNA is collected with a syringe, inserting a needle in the top of the tube to allow air to enter.

could be considered the positive control of the assay, it is possible to include other candidate genes that have to be tested for their transforming capacity. 1. Established mouse fibroblasts NIH 3T3 and 3T3 A31 are cultured in supplemented DMEM in an incubator at 37°C with 5% CO2. 2. 1 × 105 cells are plated in 100-mm tissue-culture dishes the day preceding so that they can be around 50% confluent at the transfection step (see Note 4).

The established amount of HPV-16 E7 DNA necessary to get a reasonable percentage of transformed cells is 5 to 7 µg for a 100-mm dish. A selectable marker also needs to be included in the transfection, and the neomycin gene (such as psv2neo or pCDNA3) conferring resistance to Geneticin-sulphate (G-418) is one of the choices (33). One to two micrograms of this plasmid should be added in the transfection, which is carried out using a standard calcium phosphate co-precipitation procedure (6). 1. All of the single DNA plasmids used for one-dish transfection are placed in a 1.5-mL tube containing TE to a final volume of 190 µL plus 22 µL of 2.5 M CaCl2. 2. 200 µL of this mixture are then transferred drop-wise to 200 µL of 2X HBS (Fig. 2).


387

Fig. 2. Scheme of DNA transfection. DNA is placed together with 22 µL of 2.5 M CaCl2 plus 190 µL of TE in an Eppendorf and then is mixed with 200 µL of 2X HEPESbuffered saline for 40 min. DNA is then placed on the cells, which are subjected to a glycerol shock after 4–5 h. 3. After incubation for 40 min at room temperature, the DNA mixture is added to the dish of cells containing 4.5 mL of freshly added medium. 4. After 5 h, the cells are washed with 1X Versene solution followed by serum-free medium. 5. The cells are then exposed to a solution of 14% glycerol in DMEM without serum for 1 min. After two washes with serum-free DMEM, supplemented DMEM is added back to the cells together with the appropriate concentration of the selection: 400 µg/mL of G-418. 6. The cells are then left to select over a period of 1–2 wk. Individual colonies can be selected by ring cloning (see below), or a polyclonal pool of transformed cells can be established. The readout for transformation in these assays is by assessing the ability of these cells to grow in an anchorage-independent manner. This is done in soft agar and is described below under Subheading 3.7.

3.3. Extraction and Culture of Primary Epithelial Cells 1. BMK or BRK cells are obtained by tissue extraction from kidneys of 9-d-old mice (BALB/c) or rats (Wistar Hannover), respectively (see Note 5). 2. After sacrifice, the animals are washed several times with 100% absolute ethanol to avoid any type of external contamination during cell preparation. The kidneys are removed from each animal by cutting the back until the organs are visible (Fig. 3A) (see Note 6). The kidneys, once extracted, are placed immediately in a

388

Massimi and Banks

Fig. 3. Scheme of epithelial cell extraction. Kidneys are extracted from sacrificed animals (A,B) and are homogenized and trypsinized (C,D). Cells are subsequently placed in tissue-culture dishes (E) and cultured as normal with complete Dulbecco’s modified Eagle’s medium.

3.

4.

5. 6.

50-mL Falcon tube containing serum-free supplemented DMEM (Fig. 3B) (see Note 7). The kidneys are then washed three times in serum-free supplemented DMEM to eliminate the residues of blood and adipocytes. This is also an important step, since the fat present close to the kidneys can interfere with the following trypsinization. The organs are then placed in a 60-mm dish with 2 mL of serum-free supplemented DMEM, where they are chopped with scissors into small homogeneous pieces (Fig. 3C). The material is then transferred into a fresh 50-mL tube and washed three times with serum-free DMEM, and then once with PBS. The cells are then subjected to a series of three 15-min incubations at 37°C/10% CO2 in 5 mL of 0.25% trypsin/10X Versene. At the end of each incubation, the supernatant is removed into another 50-mL Falcon tube containing 15 mL FBS, which is essential for stopping the trypsin reaction (Fig. 3D). These cells should also be kept at 37°C. At the end of the three trypsinizations, the cells are centrifuged for 10 min at 3000g. The pellet is resuspended in 10 mL of supplemented DMEM (with FBS), and, at this point, the cells are ready to be plated in 100-mm tissue-culture dishes. These should be divided between the appropriate number of dishes and cultured overnight at 37°C (Fig. 3E).


389

3.4. Transfection of Epithelial Cells After one day in culture, BRK cells are ready to be transfected. In the case of BMK it is usually necessary to leave them for an additional 24 h. Transformation of primary epithelial cells requires the cooperation of several genes (2,34), and in this context different oncogenes have been classified as establishment genes and transforming genes (1,2,35). In the case of HPV, E6 and E7 function as establishment oncogenes, while ras and fos act as transforming oncogenes. Transfection of primary epithelial cells is carried out in exactly the same way as described above for established cells (6). Typically, 5 µg of pJ4Ω HPV16 E7 or E6 would be combined together with 3 µg of EJ-ras expression plasmid plus 1–2 µg of psv2neo.

3.5. Selection of Colonies and Generation of Transformed Cell Lines 1. Following transfection and glycerol shock, the cells are placed onto selection (200 µg/mL of G-418). The medium is changed on the plates every 3–4 d and the colonies are left to develop and grow for at least 2 wk. 2. After this time, single distinct colonies are clearly visible in the HPV-16 E7 plus EJ-ras transfections. There will be essentially two types of transformed colonies with distinct morphology: one, where cells are very spikey and weakly adherent (Fig. 4A); the other, where cells are very flat and form epithelial sheets (Fig. 4B). Both of these are transformed cells. It is also possible to find some contaminations of fibroblast colonies that are clearly recognizable by their different morphology (Fig. 4C). If the experiment has been designed to compare the number of the transformed colonies obtained with different DNAs, these fibroblastic colonies have to be eliminated from the total number. 3. At this stage of the assay, the colonies can be stained to allow visualization and counting. Cells are washed with PBS and fixed with 10% paraformaldehyde/PBS for 20 min. After several washes with PBS, the dishes are stained with 10% Giemsa Blue in PBS for 1 h. Following extensive washes in water, the blue colonies are visible on the plate (Fig. 5). 4. If cell lines need to be established, then the dish is washed with PBS and single colonies are picked by trypsinization using plastic cloning rings that are commercially available. The rings are fixed on the plate using vacuum grease, and 50 µL of trypsin are placed in the ring and left to react for 5 min. 50 µL of DMEM with 10% FBS are added to stop the trypsin reaction, and the cells are then transferred to a 24-well tissue-culture plate. When confluent, the cells are passaged to 12-well and then to 6-well plates, and ultimately to a 75-cm2 flask.

3.6. Verification of Continued Oncoprotein Expression Although it is very unusual to obtain transformed colonies from BRK cells harboring both oncoproteins, BMK cells on the other hand are more susceptible to higher levels of background transformation. In the case of established

390

Massimi and Banks

Fig. 4. Different cell morphologies that could be present in a transformation assay. (A) Spiky and weakly adherent cells. (B) Flat cells which can form epithelial sheets. (C) Fibroblast cells.

Fig. 5. Staining of colonies. Cells are fixed with 10% paraformaldeyde and stained with Giemsa Blue. (A) ras alone transfection. (B) Human papillomavirus (HPV)-16 E7 + ras transfection.


391

rodent cells, the proliferation is possible without the added oncogene; therefore, in all these cases it is essential that any lines that are generated be analyzed for the continued expression of the oncoprotein of interest. The following describes such an analysis for E7, but this is relevant for any other combination of viral or cellular proteins. 1. The cells are extracted in lysis buffer and proteins are separated on SDS-PAGE gel, then electrophoretically transferred to nitrocellulose 0.22-µm membrane following the Maniatis protocol (31). 2. Membranes are blocked to avoid aspecific signal with incubation in 10% milk in PBS at 37°C for 1 h, then incubated with anti-E7 antibody diluted 1:100 in PBS with 10% milk powder and 0.5% Tween. After 2 h, the blot is washed in PBS/ 0.5% Tween and then incubated with biotinylated anti-mouse antibody diluted 1:1000 in PBS/10% milk powder/0.5% Tween for another hour. After several washes of the membrane, the final incubation is carried out with avidin-peroxidase conjugate diluted 1:1000 in PBS/10% milk/0.5% Tween. The reaction is then developed using the ECL kit, according to the manufacturer’s instructions.

3.7. Soft Agar Assays A key characteristic of transformed cells is their ability to grow in an anchorage-independent manner (36–39). This can be assessed easily by performing a soft agar assay. 1. Following the method developed by Macpherson and Montagnier (40), prepare 0.5% agar in supplemented DMEM by boiling the agar at 5% concentration in distilled water and diluting it 1/10 in supplemented DMEM at 45°C. 3 mL of this agar solution are poured into 60-mm tissue-culture dishes and left to set. The remaining agar is kept at 45°C. 2. Cells of interest are trypsinized, counted, and diluted to be at a concentration of 3 × 105/mL. One mL of this suspension is added to a 5-mL bijoux tube containing 2 mL of agar solution. This is mixed, and 0.5 mL are added to the previously prepared dishes and left to set. 3. Then the plates are incubated at 37°C overnight, and the following day the dishes are overlaid with a further 2 mL of 0.5% agar/DMEM solution. If the cells are fully transformed, then small colonies start to appear within 3–4 d. Colonies can be stained overnight with p-iodonitrotetrazolium violet (1 g/L) in PBS, photographed, and counted through an inverted microscope.

4. Notes 1. The pH is critical for the efficiency of the transfection. 2. The choice of expression plasmid obviously depends on the nature of the assays; however, efficient transformation can be obtained with pJ4Ω HPV-16 E7 (6,7,12–15), pJ4Ω HPV-16 E6 (8,16), or pJ4Ω HPV-16 E5 (9,10). The cooperating oncogene of choice is EJ-ras (11,16,17) and selectable marker (geneticin sulphate: G-418) encoded by psv2neo (6,7,12–15).

392

Massimi and Banks

3. It is also optional to amplify plasmids that replicate to only moderate copy numbers in their host bacteria (e.g., pJ4Ω), by adding 400 µL of chloramphenicol (34 mg/mL in ethanol) to a final concentration of 170 µg/mL. This treatment has the added advantage of inhibiting bacterial replication and reducing viscosity of the lysate, thereby simplifying purification of the plasmid DNA. Plasmids of a later generation (e.g., pCDNA3) replicate to such high copy numbers that amplification is unnecessary. 4. It is very important that cells are not over-confluent prior to transfection, because this could reduce transfection efficiency as well as the efficiency of the antibiotic selection. 5. The number of animals that have to be sacrificed depends on the number of cells that are required for the following experiments. In an average preparation, it is possible to extract enough cells from one rat kidney to plate three to four dishes (100 mm), which will be 50% confluent the following day. In the case of mice, assume an additional day of growth prior to transfection. 6. During the extraction, it is also important to avoid the intestine, which is situated very close to the right-hand side kidney, because this is a major source of bacterial contamination. 7. It is extremely important at this stage, as well as later, not to allow the organs to desiccate.

References 1. Land, H., Parada, L. F., and Weinberg, R. A. (1983) Cellular oncogenes and multistep carcinogenesis. Science 222, 771–778. 2. van der Eb, A. J. and Bernards, R. (1984) Transformation and oncogenicity by adenoviruses. Curr. Top. Microbiol. Immunol. 110, 23–51. 3. Heilmann, V. and Kreienberg, R. (2002) Molecular biology of cervical cancer and its precursors. Curr. Womens Health Rep. 2, 27–33. 4. Sunokawa, Y., Takebe, N., Kasamatsu, T., Terada, M., and Sugimura, T. (1986) Transforming activity of human papillomavirus type 16 DNA sequence in a cervical cancer. Proc. Natl. Acad. Sci. USA 83, 2200–2203. 5. Yasumoto, S., Burkhardt, A. L., Doniger, J., and DiPaolo, J. A. (1986) Human papillomavirus type 16 DNA-induced malignant transformation of NIH 3T3 cells. J. Virol. 57, 572–577. 6. Matlashewski, G. J., Osborn, K., Murray, A., Banks, L., and Crawford, L. V. (1987) Transformation of mouse fibroblasts with HPV type 16 DNA using a heterologous promoter. In Cancer Cells, Papillomaviruses vol. 5 (Steinberg, B.M., Brandsma, J.L., and Taichman, L. B., eds.), New York, Cold Spring Harbor, pp. 195–199. 7. Bedell, M. A., Jones, K. H., Grossman, S. R., and Laimins, L. A. (1989) Identification of human papillomavirus type 18 transforming genes in immortalized and primary cells. J. Virol. 63, 1247–1255. 8. Sedman, S. A., Barbosa, M. S., Vass, W. C., et al. (1991) The full-length E6 protein of human papillomavirus type 16 has transforming and trans-activating


9.

10.

11.

12.

13.

14.

15.

16.

17. 18.

19.

20.

21. 22.

393

activities and cooperates with E7 to immortalize keratinocytes in culture. J. Virol. 65, 4860–4866. Pim, D., Collins, M., and Banks, L. (1992) Human papillomavirus type 16 E5 gene stimulates the transforming activity of the epidermal growth factor receptor. Oncogene 7, 27–32. Leechanachai, P., Banks, L., Moreau, F., and Matlashewski, G. (1992) The E5 gene from human papillomavirus type 16 is an oncogene which enhances growth factor-mediated signal transduction to the nucleus. Oncogene 7, 19–25. Matlashewski, G., Schneider, J., Banks, L., Jones, N., Murray, A., and Crawford, L. (1987) Human papillomavirus type 16 DNA cooperates with activated ras in transforming primary cells. EMBO J. 6, 1741–1746. Crook, T., Morgenstein, J. P., Crawford, L. V., and Banks, L. (1989) Continued expression of HPV 16 E7 protein is required for maintenance of the transformed phenotype of cells co-transformed by HPV16 plus EJ-ras. EMBO J. 8, 513–519. Phelps, W. C., Yee, C. L., Munger, K., and Howley, P. M. (1988) The human papillomavirus type 16 E7 gene encodes transactivation and transformation functions similar to those of adenovirus E1A. Cell 53, 539–547. Storey, A., Pim, D., Murray, A., Osborn, K., Banks, L., and Crawford, L. (1988) Comparison of the in vitro transforming activities of human papillomavirus types. EMBO J. 7, 1815–1820. Kanda, T., Watanabe, S., and Yoshiike, K. (1988) Immortalisation of primary rat cells by human papillomavirus type 16 subgenomic DNA fragments controlled by the SV40 promoter. Virology 165, 321–325. Vousden, K. H., Doniger, J., DiPaolo, J. A., and Lowy, D. R. (1988) The E7 open reading frame of human papillomavirus type 16 encodes a transforming gene. Oncogene Res. 3, 167–175. Storey, A. and Banks, L. (1993) Human papillomavirus type 16 E6 gene cooperates with EJ-ras to immortalize primary mouse cells. Oncogene 8, 919–924. Pim, D., Storey, A., Thomas, M., Massimi, P., and Banks, L. (1994) Mutational analysis of HPV-18 E6 identifies domains required for p53 degradation in vitro, abolition of p53 transactivation in vivo and immortalisation of primary BMK cells. Oncogene 9, 1869–1876. Durst, M., Dzarlieva-Petrusevska, R. T., Boukamp, P., Fusenig, N. E., and Gissmann, L. (1987) Molecular and cytogenetic analysis of immortalized human primary keratinocytes obtained after transfection with human papillomavirus type 16 DNA. Oncogene 1, 251–256. Pirisi, L., Yasumoto, S., Feller, M., Doniger, J., and DiPaolo, J. A. (1987) Transformation of human fibroblasts and keratinocytes with human papillomavirus type 16 DNA. J. Virol. 61, 1061–1066. Kaur, P. and McDougall, J. K. (1988) Characterization of primary human keratinocytes transformed by human papillomavirus type 18. J. Virol. 62, 1917–1924. Schlegel, R., Phelps, W. C., Zhang, Y. L., and Barbosa, M. (1988) Quantitative keratinocyte assay detects two biological activities of human papillomavirus DNA and identifies viral types associated with cervical carcinoma. EMBO J. 7, 3181–3187.

394

Massimi and Banks

23. McCance, D. J., Kopan, R., Fuchs, E., and Laimins, L. A. (1988) Human papillomavirus type 16 alters human epithelial cell differentiation in vitro. Proc. Natl. Acad. Sci. USA 85, 7169–7173. 24. Pecoraro, G., Morgan, D., and Defendi, V. (1989) Differential effects of human papillomavirus type 6, 16, and 18 DNAs on immortalization and transformation of human cervical epithelial cells. Proc. Natl. Acad. Sci. USA 86, 563–567. 25. Woodworth, C. D., Doniger, J., and DiPaolo, J. A. (1989) Immortalization of human foreskin keratinocytes by various human papillomavirus DNAs corresponds to their association with cervical carcinoma. J. Virol. 63, 159–164. 26. Park, N. H., Min, B. M., Li, S. L., Huang, M. Z., Cherick, H. M., and Doniger, J. (1991) Immortalization of normal human oral keratinocytes with type 16 human papillomavirus. Carcinogenesis 12, 1627–1631. 27. Hudson, J. B., Bedell, M. A., McCance, D. J., and Laiminis, L. A. (1990) Immortalization and altered differentiation of human keratinocytes in vitro by the E6 and E7 open reading frames of human papillomavirus type 18. J. Virol. 64, 519–526. 28. Halbert, C. L., Demers, G. W., and Galloway, D. A. (1991) The E7 gene of human papillomavirus type 16 is sufficient for immortalization of human epithelial cells. J. Virol. 65, 473–478. 29. Straight, S. W., Hinkle, P. M., Jewers, R. J., and McCance, D. J. (1993) The E5 oncoprotein of human papillomavirus type 16 transforms fibroblasts and effects the downregulation of the epidermal growth factor receptor in keratinocytes. J. Virol. 67, 4521–4532. 30. Venuti, A., Salani, D., Poggiali, F., Manni, V., and Bagnato, A. (1998) The E5 oncoprotein of human papillomavirus type 16 enhances endothelin-1-induced keratinocyte growth. Virology 248, 1–5. 31. Maniatis, T., Fritsch, E. F., and Sambrook, J. (1982). Molecular Cloning, a Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. 32. Birnboim, H. C. and Doly, J. (1979) A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7, 1513–1523. 33. Southern, P. J. and Berg, P. (1982) Transformation of mammalian cells to antibiotic resistance with a bacterial gene under control of the SV40 early region promoter. J. Mol. Appl. Genet. 1, 327–341. 34. Ruley, H. E. (1983) Adenovirus early region 1A enables viral and cellular transforming genes to transform primary cells in culture. Nature 304, 602–606. 35. Rassoulzadegan, M., Naghashfar, Z., Cowie, A., et al. (1983) Expression of the large T protein of polyoma virus promotes the establishment in culture of “normal” rodent fibroblast cell lines. Proc. Natl. Acad. Sci. USA 80, 4354–4358. 36. Noda, T., Yajima, H., and Ito, Y. (1988) Progression of the phenotype of transformed cells after growth stimulation of cells by a human papillomavirus type 16 gene function. J. Virol. 62, 313–324. 37. Miyasaka, M., Takami, Y., Inoue, H., and Hakura, A. (1991) Rat primary embryo fibroblast cells suppress transformation by the E6 and E7 genes of human papillomavirus type 16 in somatic hybrid cells. J. Virol. 65, 479–482.


395

38. Hurlin, P. J., Kaur, P., Smith, P. P., Perez-Reyes, N., Blanton, R. A., and McDougall, J. K. (1991) Progression of human papillomavirus type 18-immortalized human keratinocytes to a malignant phenotype. Proc. Natl. Acad. Sci. USA 88, 570–574. 39. Leechanachai, P., Banks, L., Moreau, F., and Matlashewski, G. (1992) The E5 gene from human papillomavirus type 16 is an oncogene which enhances growth factor-mediated signal transduction to the nucleus. Oncogene 7, 19–25. 40. Macpherson, I. and Montagnier, L. (1964) Agar suspension culture for the selective assay of cells transformed by polyoma virus. Virology 23, 291–294.

Adeno-Associated Virus and HPV Interaction

397

28 Analysis of Adeno-Associated Virus and HPV Interaction Paul L. Hermonat, Hong You, C. Maurizio Chiriva-Internati, and Yong Liu Summary It is slowly becoming accepted that adeno-associated virus (AAV) is another significant factor involved in cervical carcinogenesis. However, unlike human papillomavirus (HPV), which is positively associated with cervical cancer, AAV is negatively associated with this cancer. This negative association appears to be through a direct and complex bi-directional interaction between AAV and HPV. Essentially all assays used for studying HPV can be used for studying the AAV-HPV interaction. This is because both viruses are productive in the same tissue, the stratified squamous epithelium (skin). Their relationship can be studied on the level of the complete virus and their complete life cycle using the organotypic epithelial raft culture system, which generates a stratified squamous epithelium. Their relationship can be studied in various other tissue-culture models measuring oncogenic potential. Their interaction can also be studied on the component level, as both protein–protein and protein–DNA interactions are known. Their relationship has even been studied using transgenic animals. The AAV-HPV relationship can be broken down into two halves—AAV-encoded products, which affect HPV biology, and HPV-encoded products, which affect AAV biology. To date, the former are much better studied than the latter. The rep gene and its largest product, Rep78, are responsible for most of AAV’s effects upon HPV. This chapter largely focuses on AAV’s effect on the HPV life cycle.

1. Introduction It is well known that human papillomaviruses (HPV) are positively associated with cervical cancer. Although less well known, multiple epidemiological studies substantiate that adeno-associated virus (AAV) is negatively associated with this same cancer. Through the use of the organotypic epithelial raft culture system and other assays, we now know there is a complex bidirectional interaction between AAV and papillomaviruses, with both positive and negative aspects.

From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

397

398

Hermonat et al.

1.1. Epidemiology of AAV and Cervical Cancer AAV is naturally found in the female genital tract, often in association with HPV (1–4). Multiple sero-epidemiological studies have demonstrated an inverse association between anti-AAV antibody and cervical cancer (5–7). As antibody titers reflect virus challenge, these studies are consistent with AAV infection playing a protective role against cervical cancer. A recent blinded epidemiologic study identifying AAV DNA isolated from normal and diseased cervical samples, has also confirmed the negative association of AAV with cervical carcinogenesis (odds ratio association with high-grade cervical lesions of 0.32) (8).

1.2. Early Studies on AAV Interaction With Bovine Papillomavirus and Identification of AAV Rep78 as the Antipapillomavirus Protein The possibility of an AAV–papillomavirus interaction was first investigated using bovine papillomavirus (BPV) in C127 mouse fibroblasts as a model. In these early experiments, AAV Rep78 was found to both inhibit BPV-induced oncogenic transformation and BPV DNA replication (9–11). Chimeric BPVAAV genomes, which include the Rep78 coding sequences, were also found to be defective in both of these activities (12).

1.3. Later Studies on AAV Interaction With HPV Once assays were developed for studying HPV-induced oncogenic transformation, AAV was found to have a similar inhibitory affect on HPV 16 and 18 (13–15). AAV was also found to inhibit HPV in animal models, including transgenics (16,17). It is the Rep78 protein, encoded by the AAV rep gene, which is responsible for this inhibition (9,13–15). Thus, the study of AAV Rep78, an “anti-oncogene analog,” and its mechanism of action, is a unique opportunity and perspective from which to study the molecular origins of cancer. Rep78 is known to affect HPV through its interaction with papillomavirus DNA, its encoded proteins, and a variety of relevant cellular proteins (18–25). Figure 1 shows the inhibition of HPV-16-directed oncogenic transformation of C127 cells by AAV Rep78 expression. Figure 2 shows a compilation of Rep78’s published interactions with papillomavirus proteins and DNA. Figure 3 demonstrates the binding of AAV-encoded Rep78 to the HPV-16 p97 promoter region. AAV also benefits from HPV. AAV, like HPV, is productive in skin (26), but AAV replication is enhanced by HPV (27,28). New advances are expected in the study of AAV–HPV interaction in the organtypic raft culture system, a model of squamous epithelium, as both viruses are productive in this tissue, it being their natural host. The latest results from such studies indicate that the presence of AAV temporally accelerates the HPV life cycle (29).


399

Fig. 1. Inhibition of human papillomavirus (HPV)-16-directed oncogenic transformation of contact-inhibited C127 cells by adeno-associated virus (AAV) Rep78. Eighty percent confluent plates of C127 cells were transfected with 4 µg of pL67R (HPV 16/ras) plus 8 µg of an AAV plasmid where indicated. After growth for 3 wk, the cells were fixed with formaldehyde and stained with methylene blue. Note that wildtype Rep78 inhibited foci formation, whereas mutant Rep78 did not.

Fig. 2. Established direct interactions of adeno-associated virus Rep78 with human papillomavirus.

Fig. 3. Binding of Rep78 to the p97 region of human papillomavirus (HPV) 16. The indicated amounts of MBP-Rep78 and 32P-labeled p97 (nt 14-106) were incubated together, and then the sample was electrophoresed on a nondenaturing polyacrylamide gel. Note that a protein–DNA complex (indicated by the arrow) forms in a dosagedependent manner for Rep78.

400

Hermonat et al.

2. Materials 1. pAT–HPV 16 (available from Dr. Richard Schlegel, BSB/113, Pathology, Georgetown University Medical Center, 3900 Reservoir Rd., NW, Washington, DC 20057, e-mail [email protected]). 2. pL67R (available from Dr. Paul L. Hermonat, address above). 3. pMal-Rep78 (MBP-Rep78) (available from Dr. Paul L. Hermonat, address above). 4. pKEX-Rep78 (available from Dr. Jurgen Kleinschmidt, Angewandte Tumorvirologie, Deutsches Krebsforschungszentrum, Im Neuenheimer Feld 242, 69120 Heidelberg, Germany, e-mail: [email protected]). 5. pSM620 (available from Dr. Paul L. Hermonat, address above). 6. p97 CAT (available from Dr. Peter M. Howley, Pathology, Rm 630, Harvard Medical School, 200 Longwood Ave., Boston, MA 02115, e-mail [email protected]). 7. T-47D cell nuclear extract (Geneka Corp., HPV-negative breast ductal carcinoma). 8. Protein Purification and Expression System (New England Biolabs). 9. HeLa cell nuclear extract (Geneka Corp., HPV-positive cervical carcinoma). 10. Primary human foreskin keratinocytes (obtained from clinics). 11. CIN612 9E cells (available from Dr. Craig Meyers, Department of Microbiology and Immunology, Pennsylvania State University College of Medicine, 500 University Drive, Hershey, PA 17033, e-mail [email protected]). 12. Keratinocyte serum-free medium (Gibco/BRL). 13. Dulbecco’s modified Eagle’s medium (DMEM). 14. LipofectACE (Gibco/BRL). 15. Binding buffer: 25 mM HEPES KOH (pH 7.5), 10 mM MgCl 2, 1 mM dithiothreitol (DTT), 2% glycerol, 25 µg bovine serum albumin, 50 mM NaCl, 0.01% NP40, and 0.5 µg poly (dI-dC) per ng of DNA. 16. Heparin-agarose (Sigma H6508). 17. Column-washing buffer: 0.254 M NaCl-phosphate-buffered saline (PBS) (pH 7.4). 18. Column elute buffer: 0.554 M NaCl-PBS (pH 7.4). 19. In vitro transcription reaction mixture: 0.5 µg of DNA template; 20 mM HEPES (pH 7.9); 5 mM MgCI2; 100 mM KCl; 0.5 mM DTT; 20% glycerol; 25 µM [32P] GTP, 400 µM ATP, CTP, and UTP; and 8 U of chosen nuclear extract in a total volume of 25 µL. 20. Cesium chloride. 21. Stop solution: 300 mM Tris HCl (pH 7.9), 0.5% sodium dodecyl sulfate (SDS), 300 mM sodium acetate, 2 mM ethylenediamine tetraacetic acid (EDTA), and 3 µg/mL tRNA. 22. 10-cm Tissue-culture plates. 23. Fetal bovine serum. 24. 1 mM Calcium chloride. 25. 0.5 µg/mL Hydrocortisone. 26. 293 Cells (ATCC). 27. Adeno-associated virus (AAV, available from ATCC).

Adeno-Associated Virus and HPV Interaction 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45.

401

Adenovirus type 2 (available from ATCC). DNase I. Deoxycholic acid (Sigma). 2.5-cm-Diameter glass column (Sigma C4669) that includes a Luer Lock (Sigma S7396). Filtration membrane (Sigma S7271). Biomax-100K NMWL-filter device (Millipore). Centricon 10-kDa cut-off membrane filters (Amicon). HPV 31b virus. Ultrafree-15 centrifugal filter device with 10K NMWL membrane (Millipore). Primer 1: 5'ACAAGCAGGATTGAAGGCCA, HPV 16 nt 7043-7065. Primer 2: 5'CATATCACCAGC TCACCGTC, nt 615-633 of pSV2CAT. Phenol-chloroform. Ethanol. Formamide, 0.1% xylene cyanol, 0.1% bromophenol blue. 6% Polyacrylamide, 7 M urea gel. 5% Glycerol. 0.5X TBE buffer: 45 mM Tris-borate, 1 mM EDTA. 32P ATP (5000 Ci/mmol, Amersham).

3. Methods (see Note 1) 3.1. Assays for Studying the Effects of AAV and Its Rep78 Gene on Inhibition of HPV-Induced Oncogenic Transformation

3.1.1. Focus Formation Assays A wide variety of related focus formation assays are available, and some of these are covered in Chapter 27. The differences in these assays are the specific HPV plasmid being used, the specific contact-inhibited cell line used, whether G418 selection is needed for the enrichment of transfected cells, and the length of culture time needed to allow for oncogenic foci to appear. These assays can be modified to study AAV-HPV interaction by the inclusion of an AAV plasmid. Appropriate AAV-containing plasmids would include the cloned wild-type AAV plasmid pSM620 (30) or the AAV Rep78 expression plasmid pKEX-Rep78 (14). Usually in these assays a 2-to-1 ratio of AAV plasmid to HPV plasmid is used.

3.1.2. Transformed Keratinocyte Outgrowth Assay Another assay related to focus formation is the transformed keratinocyte outgrowth assay (15). In this assay, oncogenically transformed keratinocytes will outgrow normal primary keratinocytes and fibroblasts when grown in calcium-supplemented DMEM. To more easily observe oncogenic transformation, a chimeric HPV-16/c-H-ras plasmid was generated in which the p97 promoter and the E6 and E7 genes were left intact and the E1 coding sequence

402

Hermonat et al.

was replaced with the coding sequences of c-H-ras, with ras expressed from the p97 promoter. Thus, this plasmid contained three oncogenes; it was called pL67R (representing LCR, E6, E7, ras). 1. Grow cultures of normal primary human foreskin keratinocytes in keratinocyte serum-free medium, to 80% confluence in 10-cm plates. 2. Transfect these primary cells with 4 µg of pL67R (HPV-16 E6, E7, ras plasmid) plus 8 µg of an AAV plasmid such as pSM620 (wild-type AAV) or pKEX-Rep78, using LipofectACE as specified by the manufacturer. 3. At 1 wk posttransfection, replace the medium with 50% keratinocyte serum-free medium/50% DMEM with 7% fetal bovine serum, 1 mM CaCl, and 0.5 µg/mL hydrocortisone. 4. At 4 and 5.5 wks posttransfection, split the cells 1:4. 5. At 7 wk, formalin fix and methylene blue stain the plates. The oncogenically transformed keratinocytes have a growth advantage over fibroblasts and non-transformed keratinocytes, and eventually replace these cells. The oncogenically transformed cells stain as very dark and dense areas. Quantitate the percent coverage of HPV and HPV+AAV-treated plates and compare them to each other and the null control. Using duplicate plates, the HPV and AAV DNA sequences can be identified by isolating total cellular DNA, digesting the DNA with appropriate restriction enzymes, size separating the DNA by agarose gel electrophoresis, and probing with radio-labeled AAV or HPV sequences.

3.1.3. Cell Growth in Soft Agar Tumor cell growth in soft agar is another standard assay system that can be used to study AAV inhibition of HPV, and this assay is covered in Chapter 27. This assay can be modified by the inclusion of an AAV plasmid, such as the wild-type AAV plasmid pSM620 (30) or the AAV Rep78 expression plasmid pKEX-Rep78 (14).

3.1.4. Tumor Growth in Mice There are a variety of tumor growth model systems in animals (e.g., see Chapter 16) that be used to study AAV inhibition of HPV, but these are too extensive to be covered in this chapter. These assays can be modified by the inclusion of an AAV plasmid, such as the wild-type AAV plasmid pSM620 or the AAV Rep78 expression plasmid pKEX-Rep78.

3.2. Effect of AAV and Rep78 on the HPV Life Cycle (see Notes 2–4) 3.2.1. Studying AAV’s Effect on HPV Replication During Productive Infection, in the Organotypic Epithelial Raft Culture System Both HPV and AAV are productive in the organotypic epithelial raft culture system (26,28,29). Thus, this system will be a central and important assay for


403

much future research and is described in detail in Chapters 12–14. An important point to address here is that HPV is a relatively slow virus, taking 10–12 d of culture to produce progeny virus. In sharp contrast, AAV is relatively fast, completing its life cycle in approx 3–5 d. Moreover, when the two viruses are together, the presence of AAV stimulates the more rapid completion of the HPV life cycle, reducing it from 10–12 d to 4–6 d (29). The standard raft assay can be modified by infecting HPV-positive keratinocytes with wild-type AAV. Alternatively, these cells can be transfected with an AAV plasmid using a lipofection agent. Studying many variables is possible, e.g., the temporal introduction of the viruses. Primary keratinocytes could be infected with AAV first before the introduction/infection of HPV. Changes in viral DNA levels can then be analyzed by isolating total cellular DNA, size separating the DNA by agarose gel electrophoresis, Southern blotting the DNA, and probing with appropriate sequences. 3.2.1.1. GENERATION OF PURIFIED AAV VIRUS STOCK 1. Infect 293 cells with AAV and adenovirus type 2, both at multiplicity of infection of 5. 2. At 36 h, freeze-thaw the cells three times and column purify the AAV virus particles by the technique described by Auricchio et al. as below (34). 3. Treat AAV virus solution with DNase I (10 U/mL for 1 h at 37°C). 4. Treat virus solution with a final concentration of 0.5% deoxycholic acid for 30 min at 37°C. 5. Prepare column by pipetting 8 mL of heparin-agarose suspension into a 2.5-cmdiameter glass column that includes a Luer lock. 6. After flow through, apply a filtration membrane on top of the agarose bed. 7. Equilibrate the column with 25 mL of column-washing buffer. 8. Close the lock. Apply the AAV virus solution onto the heparin-agarose column and open the lock to allow 1 drop/s. 9. After loading the virus solution, wash the column with 50 mL of column-washing buffer. 10. Elute the AAV virus with 15 mL of column eluate buffer. 11. If needed, concentrate the AAV virus to approx 1 mL using a Biomax-100K NMWL filter device. 12. Assess the purity of the viral preparation (100 µL) on a 4–20% SDS-polyacrylamide gel, detecting the proteins by Coomassie staining. The presence of only VP1 (62 kDa), VP2 (73 kDa), and VP3 (87 kDa) indicates viral purity. 13. Titer purified virus by dot blot comparison. Treat 100 µL of virus stock with proteinase K at 50 µg/mL for 1 h at 37°C. This will release the encapsidated DNA. 14. Precipitate the DNA by adding 10 µg of carrier tRNA, NaCl to a final concentration of 0.1 M, and adding 2.5 vol of ethanol.

404

Hermonat et al.

15. After centrifugation, re-solubilize the DNA pellet in 5 µL of water, then add 10 µL of 0.4 N NaOH, 0.5 M NaCl for 10 min. Also prepare DNA standards of AAV DNA containing 106, 107, 108, and 109 copies of pSM620 plasmid DNA, and denature them in a similar manner. 16. Neutralize the samples with 200 µL of 0.5 M Tris-HCl, 1 M NaCl (pH 7.2) and immediately add to a dot blot apparatus, under suction, containing a pre-wetted nylon membrane. 17. Probe the dot blot membrane with 32P-labeled pSM620 DNA under standard conditions, wash with 1X SSC, and autoradiograph. 18. Compare the signal of the virus stock to the known standards to obtain a titer in encapsidated genomes per mL.

3.2.1.2. GENERATION OF PURIFIED HPV-31B VIRUS STOCK

HPV-31b virus stock can be generated from mature organotypic epithelial cell rafts generated using CIN612 9E cells, abbreviated to “612” hereafter, as described by You et al. (35). These cells harbor a wild-type, fully active HPV31b genome at approx 50 copies per cell, and can be rafted. 1. Twelve days after raising the raft to the air interface, harvest the raft tissues, mince, and homogenize in DMEM. 2. Treat the “rough” virus stocks with DNase I and titer it by dot blot hybridization to give the number of encapsidated genomes per mL. 3. For further purification, a cesium chloride (CsCl) gradient can be used. Add CsCl to the rough virus stock at the density of 1.3 g/mL. Form a gradient by ultracentrifugation at 135,000g for 24 h at 4°C. To harvest the fractions, puncture the tube and collect 1-mL fractions from the bottom. 4. Use 0.1 mL of each fraction and 0.1 mL of the “rough” cell lysate (without CsCl purification) for harvesting DNA, and analyze the amount of HPV DNA by Southern blotting and hybridization with 32P-HPV-31b DNA probe. 5. Dialyze the highest HPV-31b virus-containing fraction using an Ultrafree-15 centrifugal filter device with 10K NMWL membrane. After removal of the initial salt by three wash cycles with PBS at 2000 g, centrifuged at 4°C, store the purified HPV-31b virus stock at –80°C.

3.2.2. Effect of AAV/Rep78 on Regulation of HPV Transcription 3.2.2.1. STUDYING AAV REP78’S EFFECT ON THE HPV-16 P97 PROMOTER TRANSIENT CHLORAMPHENICOL ACETYLTRANSFERASE ASSAY

BY

Transient chloramphericol acetyltransferase (CAT) assays are a wellaccepted standard for studying promoter regulation. However, an often overlooked point is that this assay really includes both transcriptional and translational regulation. This assay is described in more detail in Chapter 22. Briefly, carry out transient CAT assays as follows: 1. Calcium phosphate transfect the p16P (p97-CAT) plasmid (4 µg) (36) plus the wild-type cloned-AAV plasmid pSM620 (30) or the Rep78 expression plasmid


405

pKEX-Rep78 plasmid (8 µg of either plasmid) (14). Variable amounts of CAT and AAV plasmid can be used, but usually twice as much AAV plasmid is transfected. 2. Forty-eight hours after transfection, cell extracts are prepared, equalized for protein content by spectroscopic analysis at 280 nm, and assayed for CAT activity. Alternatively, a marker expression plasmid can be co-transfected with the CATcontaining plasmid to determine transfection efficiency and allow for differences in transfection efficiency, if one’s transfection efficiency is found to be significantly variable.

3.2.2.2 STUDYING AAV REP78’S EFFECT ON THE HPV-16 P97 PROMOTER, BY IN VITRO TRANSCRIPTION IN NUCLEAR EXTRACTS

This is another standard assay for studying transcriptional regulation that has the advantage of allowing the addition of specific well-defined proteins, usually transcription factors, for the study of specific effects. 1. For studying the p97 promoter, an HPV-16 p97 CAT DNA fragment can be used as a template for transcription. Generate the p97 CAT-DNA fragment by standard PCR amplification using primer 1, complementary to the p97 sequences, and primer 2, complementary to the CAT sequences. Use the plasmid p16P (p97CAT) as the PCR template (36). The product will be 1.2 kb. 2. Carry out the in vitro transcription in a 25-µL reaction mixture with 8 U HeLa nuclear extract or 5 µg of T-47D nuclear extract. Incubate the reactions at 30°C for 60 min, and then terminate by adding 175 µL of stop solution. 3. Extract the RNA with phenol chloroform, precipitate with ethanol, and finally dissolve in 10 µL of formamide containing 0.1% each of xylene cyanol and bromophenol blue. 4. Analyze the samples on a 6% polyacrylamide, 7 M urea gel by autoradiography. The p97-specific RNA product is of approx 300 bases.

3.2.2.3. STUDYING AAV REP78’S INTERACTION WITH THE HPV-16 P97 DNA BY ELECTROPHORETIC MOBILITY SHIFT ASSAY (SEE NOTE 5)

Rep78 is known to bind to the long control region (LCR)/upstream regulatory region of papillomaviruses, and the electrophoretic mobility shift assay (EMSA) is a standard assay for observing such protein–DNA interaction. 1. A purified Rep78 protein must be generated using pMAL-Rep78 (37). Perform the purification of MBP-Rep78 using the Protein Purification and Expression System, following the kit directions. 2. Collect fractions and analyze them by SDS-polyacrylamide gel electrophoresis. 3. Concentrate the purified fractions using Centricon 10-kDa cut-off membrane filters. 4. Routinely, these procedures result in MBP-Rep78 protein of 70–90% purity with a yield of 200 µg/L bacterial culture. 5. Various regions of interest of the HPV-16 LCR may be used as protein-binding substrates.

406

Hermonat et al.

6. The sequences of interest, once generated as a synthetic double-stranded substrate, should be 5' end labeled with polynucleotide kinase using 32P ATP (5000 Ci/mmol). EMSA assays can be conducted as described in chapter 20. In brief, to carry out the EMSA assay, approx 1 ng of 5' end labeled DNA substrate should be incubated with increasing amounts of MBP-Rep78 for 10 min at room temperature in binding buffer. 7. Electrophorese the incubated products on a 4% polyacrylamide gel (40:1 acrylamide and bis-acrylamide weight ratio) with 5% glycerol in 0.5X TBE buffer at 100 V for approx 3 h. 8. After running the gel, it should be dried and autoradiographed at –70°C.

Converse to studying AAV’s affect on HPV is the analysis of HPV’s affect on AAV. While AAV is known to be fully productive in the organotypic epithelial raft culture system (26), HPV is also known to significantly enhance AAV replication (27,28). The organotypic epithelial raft culture system is very appropriate for studying HPV’s effects on the AAV life cycle. Again, AAV is the faster replicating virus in this system, so the times of harvest should be day 5 or before. Many experimental variables can be tried. AAV can be introduced either into HPV-positive keratinocytes or into normal keratinocytes transfected with an HPV subgenomic construct. Yet another variation can be dual infection by both AAV and HPV virus particles. Changes in viral DNA levels can then be analyzed by isolating total cellular DNA, size separating the DNA by agarose gel electrophoresis, and Southern blotting. 4. Notes 1. Any assay used to study HPV can be altered to study AAV’s effects on HPV. Thus, the full range of assays is not discussed here. 2. The main AAV effector is the Rep78 protein, the largest one of four proteins encoded by AAV’s rep gene. This has been determined by both the Hermonat and Kleinschmidt laboratories (11,13,14). A good choice of plasmid for expressing Rep78 is pKEX-Rep78 (Dr. Kleinschmidt), as it is mutated so that the smaller Rep proteins will not be expressed. 3. AAV Rep78 has the capability to interact with both DNA and proteins (18,19). 4. If you are observing AAV–HPV interaction in the organotypic raft culture system, be aware that wild-type AAV also replicates in this culture system. 5. For studying AAV Rep78’s in vitro biochemical interaction with HPV proteins and DNA, the use of the purified maltose-binding protein (MBP)-Rep78 fusion protein produced from the plasmid pMal-Rep78 in bacteria is recommended.

References 1. Han, L., Parmley, T. H., Keith, S., Kozlowski, K. J., Smith, L. J., and Hermonat, P. L. (1996) High prevalence of adeno-associated virus (AAV) type 2 rep DNA in cervical materials: AAV may be sexually transmitted. Virus Genes 12, 47–52.


407

2. Walz, C., Deprez, A., Dupressoir, T., Durst, M., Rabreau, M., and Schlehofer, J. R. (1997) Interaction of human papillomavirus type 16 and adeno-associated virus type 2 co-infecting human cervical epithelium. J. Gen. Virol. 78, 1441–1452. 3. Malhomme, O., Dutheil, N., Rabreau, M., Armbruster-Moraes, E., Schlehofer, J. R., and Dupressoir, T. (1997) Human genital tissues containing DNA of adeno-associated virus lack DNA sequences of the helper viruses adenovirus, herpes simplex virus or cytomegalovirus but frequently contain human papillomavirus DNA. J. Gen. Virol. 78, 1957–1962. 4. Walz, C. M., Anisi, T. R., Schlehofer, J. R., Gissmann, L., Schneider, A., and Muller, M. (1998) Detection of infectious adeno-associated virus particles in human cervical biopsies. Virology 247, 97–105. 5. Mayor, H. D., Drake, S., Stahmann, J., and Mumford, D. M. (1976) Antibodies to adeno-associated satellite virus and herpes simplex in sera from cancer patients and normal adults. Am. J. Obstet. Gyn. 126, 100–105. 6. Georg-Fries, B., Biederlack, S., Wolf, J., and zur Hausen, H. (1984) Analysis of proteins, helper dependence, and seroepidemiology of a new human parvovirus. Virology 134, 64–71. 7. Smith, J. S., Herrero, R., Erles, K., et al. (2001) Adeno-associated virus seropositivity and HPV-induced cervical cancer in Spain and Columbia. Internat. J. Cancer 94, 520–527. 8. Coker, A. L., Russell, R. B., Bond, S. M., et al. (2001) Adeno-associated is associated with lower risk of high grade cervical squamous intraepithelial lesions. Exper. Molec. Pathol. 70, 83–89. 9. Hermonat, P. L. (1989) The adeno-associated virus Rep78 gene inhibits cellular transformation induced by bovine papillomavirus. Virology 172, 253–261. 10. Schmitt, J., Schlehofer, J. R., Mergener, K., Gissman, L., and zur Hausen, H. (1989) Amplification of bovine papillomavirusDNA by N-methyl-N-nitro-Nnitrosoquanidine, ultraviolet irradiation, or infection with herpes simplex virus. Virology 172, 253–261. 11. Hermonat, P. L. (1992) Inhibition of bovine papillomavirus plasmid DNA replication by adeno-associated virus. Virology 189, 329–333. 12. Hermonat, P. L., Meyers, C., Parham, G. P., and Santin, A. D. (1998) Inhibition/ stimulation of bovine papillomavirus by adeno-associated virus is time as well as multiplicity dependent. Virology 247, 240–50. 13. Hermonat, P. L. (1994) Adeno-associated virus inhibits human papillomavirus type 16: a viral interaction implicated in cervical cancer. Cancer Res. 54, 2278–2281. 14. Horer, M., Weger, S., Butz, K., Hoppe-Seyler, F., Geisen, C., and Kleinschmidt, J. A. (1995). Mutational analysis of adeno-associated virus Rep protein-mediated inhibition of heterologous and homologous promoters. J. Virol. 69, 5485–5496. 15. Hermonat, P. L., Plott, R. T., Santin, A. D., Parham, G. P., and Flick, J. T. (1997) The adeno-associated virus Rep78 gene inhibits oncogenic transformation of primary keratinocytes by a human papillomavirus type 16-ras chimeric. Gyn. Oncol. 66, 487–494.

408

Hermonat et al.

16. Su, P. F. and Wu, F. Y. (1996) Differential suppression of the tumorigenicity of HeLa and SiHa cells by adeno-associated virus. Brit. J. Can. 73, 1533–1537. 17. Walz, C. M., Correa-Ochoa, M. M., Muller, M., and Schlehofer, J. R. (2002) Adeno-associated virus type 2-induced inhibition of the human papillomavirus type 18 promoter in transgenic mice. Virology 293, 172–181. 18. Zhan, D.-J., Santin, A. D., Parham, G. P., Li, C., Meyers, C., and Hermonat, P. L. (1999) Binding of the human papillomavirus type 16 p97 promoter by adenoassociated virus (AAV) Rep78 major regulatory protein correlates with inhibition. J. Biol. Chem. 274, 31,619–31,624. 19. Hermonat, P. L., Santin, A. D., and Zhan, D.-J. (2000) Binding of the human papillomavirus type 16 E7 oncoprotein and the adeno-associated virus Rep78 major regulatory protein in vitro and in yeast, and the potential for downstream effects. J. Hum. Virol. 3, 113–124. 20. Marcello, A., Massimi, P., Banks, L., and Giacca, M. (2000) Adeno-associated virus type 2 rep protein inhibits human papillomavirus type 16 E2 recruitment of the transcriptional coactivator p300. J. Virol. 74, 9090–9098. 21. Chon, S. K., Rim, B. M., and Im, D. S. (1999) Adeno-associated virus Rep78 binds to E2-responsive element 1 of bovine papillomavirus type 1. IUBMB Life 48, 397–404. 22. Hermonat, P. L., Santin, A. D., and Batchu, R. B. (1996) The adeno-associated virus Rep78 major regulatory, transformation suppressor protein binds cellular Sp1 and evidence of a biological effect. Cancer Res. 56, 5299–5304. 23. Hermonat, P. L., Santin, A. D., Batchu, R. B., and Zhan, D. J. (1998) The adenoassociated virus Rep78 major regulatory protein binds the cellular TATA-binding protein, TBP. Virology 245, 120–127. 24. Prasad, C. K., Meyers, C., Zhan, D. J., et al. (2003) The adeno-associated virus major regulatory protein Rep78-c-Jun-DNA motif complex modulates AP-1 activity. Virology 314, 423–431. 25. Su, P. F., Chiang, S. Y., Wu, C. W., and Wu, F. Y. (2000) Adeno-associated virus major rep78 protein disrupts binding of TATA-binding protein to the p97 promoter of human papillomavirus type 16. J. Virol. 74, 2459–2465. 26. Meyers, C., Mane, M., Kokorina, N., Alam, S., and Hermonat, P. L. (2000) Ubiquitous adeno-associated virus type 2 replicates in a model of normal skin. Virology 272, 338–346. 27. Ogston, P., Raj, K., and Beard, P. (2000) Productive replication of adeno-associated virus can occur in human papillomavirus type 16 (HPV-16) episome-containing keratinocytes and is augmented by the HPV-16 E2 protein. J. Virol. 74, 3494–3504. 28. Meyers, C., Alam, S., Mane, M., and Hermonat, P. L. (2001) Altered biology of adeno-associated virus type 2 and human papillomavirus during dual infection of natural host tissue. Virology 287, 30–39. 29. Agrawal, N., Mane, M., Chiriva-Internati, M., Roman, J. J., and Hermonat, P. L. (2002) Temporal acceleration of the human papillomavirus life cycle by adeno-


30.

31.

32. 33.

34.

35. 36.

37.

409

associated virus (AAV) type 2 super-infection in natural host tissue. Virology 297, 203–210. Samulski, R. J., Srivastava, A., Berns, K. I., and Muzyczka, N. (1983) Rescue of adeno-associated virus from recombinant plasmids: gene correction within the terminal repeats of AAV. Cell 33, 135–143. Walz, C. M., Correa-Ochoa, M. M., Muller, M., and Schlehofer, J. R. (2002) Adenoassociated virus type 2-induced inhibition of the human papillomavirus type 18 promoter in transgenic mice. Virology 293, 172–181. Su, P. F. and Wu, F. Y. (1996) Differential suppression of the tumorigenicity of HeLa and SiHa cells by adeno-associated virus. Brit. J. Cancer 73, 1533–1537. Selvakumar, R., Ahmed, R., and Wettstein, F. O. (1995) Tumor regression is associated with a specific immune response to the E2 protein of cottontail rabbit papillomavirus. Virology 208, 298–302. Auricchio, A., Hildinge,r M., O’Connor, E., Gao, G. P., and Wilson, J. M. (2001) Isolation of highly infectious and pure adeno-associated virus type 2 vectors with a single-step gravity-flow column. Hum. Gene Ther. 12, 71–76. You, H., Liu, Y., Agrawal, N., et al. (2003) Infection, replication and cytopathology of human papillomavirus type 31 in trophoblasts. Virology 316, 281–289. Romanczuk, H., Thierry, F., and Howley, P. M. (1990) Mutational analysis of cis elements involved in E2 modulation of human papillomavirus type 16 P97 and type 18 P105 promoters. J. Virol. 64, 2849–2859. Batchu, R. B., Miles, D. A., Rechtin, T. M., Drake, R. R., and Hermonat, P. L. (1995) Cloning, expression and purification of full length Rep78 of adeno-associated virus as a fusion protein with maltose binding protein in Escherichia coli. Biochem. Biophys. Res. Comm. 208, 714–720.

Substrate Degradation Induced by HPV E6 Oncoproteins

411

29 In Vitro Assays of Substrate Degradation Induced by High-Risk HPV E6 Oncoproteins Miranda Thomas and Lawrence Banks Summary The high-risk mucosal human papillomavirus E6 proteins were the first viral proteins that were shown to use the ubiquitin proteasome pathway for the inactivation of their cellular target proteins. The first substrate to be identified was the p53 tumor suppressor protein, and since then many other substrates for E6-induced degradation have been described. All of these require the presence of high-risk mucosal E6 together with the E1, E2, and E3 enzymes of the ubiquitin pathway. This activity of E6, although complex, is nonetheless amenable to in vitro analysis. Many different protocols have been described over the years for performing these assays. In this chapter we describe the most easily used and robust procedure that is routinely used in our laboratory.

1. Introduction Many cellular processes involve the controlled destruction of proteins: a good example is the degradation of cyclins used as signaling for progression through the cell cycle (1,2). A number of viruses use the proteasomal mechanisms of their host cell to alter the cellular environment to favor their own replication (3). The points where viruses impinge on cellular pathways are interesting, both for understanding how the virus replicates and, perhaps, for understanding how to stop it from doing so. Because these are complicated processes, in vitro assays are an extremely useful tool for unraveling them, and for providing potential links in the pathways that can then be further assessed by in vivo assays. 2. Materials 1. TNT coupled in vitro transcription/translation system (Promega). These kits use either rabbit reticulocyte lysate or wheat germ extract as a translation medium, and have a comprehensive set of instructions for their use (see Note 1). From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

411

412

Thomas and Banks

2. RNase-free sterile de-ionized water. 3. RNase inhibitor (Ambion). 4. Radiolabeled amino acid: [35S]-labeled methionine (Met) or cysteine (Cys), or [14C]-labeled leucine (Leu) (see Note 2). 5. In vitro expression plasmids (see Note 3 for selection, Note 4 for preparation, and Note 5 for storage). 6. TE buffer: 10 mM Tris-HCl (pH 8.0), 1 mM ethylenediaminetetraacetic acid (EDTA) 7. E1A buffer: 250 mM NaCl, 50 mM HEPES (pH 7.0), 0.1% NP40. 8. Sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) gel loading buffer: 160 mM Tris HCl (pH 6.8), 1.6% SDS, 16 mM β-mercaptoethanol, 50% glycerol. 9. Antibody reactive with the target protein of the assay. 10. Protein A-Sepharose or protein G-Sepharose in E1A buffer. 11. Solutions and apparatus for SDS-PAGE separation of protein products. 12. X-ray film. 13. Sterile RNase-free microfuge tubes. 14. Sterile RNase-free pipet tips.

3. Methods 3.1. In Vitro Translation Before starting the translation, check the amino acid sequence of your proteins to determine which radiolabeled amino acid is the most appropriate (see Notes 2 and 6). 1. In vitro translate your proteins using the TNT kit (see Notes 7 and 8). 2. Run 1 µL of each translation reaction on SDS-PAGE, dry the gel, and expose on film or on a phosphorimager. 3. Analyze the success of your translation and decide on the amounts of each translation product to use (see Note 9).

3.2. In Vitro Degradation The proportions of target protein to human papillomavirus (HPV) E6 protein need to be determined empirically, as does the optimal assay time. For example, we would usually suggest a ratio of one p53 to three E6 (as determined from the radiolabel by phosphorimager or by film), using 0-, 30-, and 60-min time points. For Dlg, which is less effectively degraded in the presence of E6, a ratio of 1:5 and time points of 1 and 2 h would be better (see Fig. 1 and Note 10). 1. First calculate the volume of each time-point sample. In the example shown in Table 1, we would recommend 6.5 to 6.8 µL each. 2. Set up reaction tubes on ice (see Note 11), using water-primed lysate (wpl) to equalize the volumes. Do not use any additional buffers (see Note 12). Shown in Table 1 is a representative assay. To reduce variation from pipetting errors, each


413

Fig. 1. A cartoon representing the variation in the kinetics of E6-induced in vitro degradation (7–9). Table 1 Example of a Degradation Assay Set-Up Loading control

A䉳 B䉳 C䉳

Tube 1 2 3 4 5 6 7

Target

}

p53 6 µL

} }

p53 4 µL p53 4 µL

E6

} }

– – – 16E6 11 µL 18E6 9 µL

WPL*

}

16.5 µL – –

}

2 µL

Time (min) at 30°C 0 30 60 30 60 30 60

*Water-primed lysate. The in vitro translated proteins are mixed in the bold assay tubes (3, 5, and 7 here). 1-µL samples are removed for input controls (lettered tubes). Zero time-point sample is taken into tube 1 and frozen. Assay tubes are placed at 30°C. Samples are taken at 30 min into tubes 2, 4, and 6, and then frozen. Samples are taken at 60 min into fresh tubes labeled 3, 5, and 7, and frozen. Bold assay tubes are discarded as radioactive waste.

3. 4. 5. 6.

reaction (e.g., p53 alone for three time points) should be set up in a single tube and the time points removed from it. The bold tube numbers are the reaction tubes; the last time point is removed into a fresh tube and any remaining reaction mixture discarded as radioactive waste. Take 1 µL of each reaction mixture into 3 µL of gel loading buffer for loading controls. Either run these immediately on SDS-PAGE gel (use 15% polyacrylamide to see E6 inputs) or hold on dry ice (or –80°C) and run them together with the assay samples. Take your zero time-point sample into tube number 1, hold on dry ice (or –80°C). Put reaction tubes at 30°C. Take samples (see Note 13) into labeled chilled tubes at the relevant time points; hold each on dry ice (or –80°C). Discard reaction tubes after taking the final time points. These will probably have some residue of reaction mixture. This is preferable to finding that you do not

414

Thomas and Banks

have enough for a complete last time-point sample: some liquid always adheres to the outside of a pipet tip. 7. Having frozen your final samples on dry ice or at –80°C, they can be stored until required, or thawed and processed immediately. 8. Run the samples from tubes A, B, and C in Table 1 (the loading controls) on a 15% SDS-PAGE gel along with the final assay samples, dry, and expose the gel to X-ray film.

3.3. Immunoprecipitation Immunoprecipitation is used simply to separate the target protein from the reaction mixture and thus to produce a cleaner result. Reticulocyte lysate is a very highly concentrated protein solution, and the radiolabel can stick to the hemoglobin, thus making very dirty-looking SDS-PAGE gels and autoradiographs; this also makes the analysis of the results much more difficult (see Note 12). Immunoprecipitation is obviously not necessary if your translations were very efficient and thus your time-point samples are very small, e.g., up to 3 µL; in this case the background smear is negligible and the samples can be run on a minigel with a great saving of time and reagents. 1. Thaw your time-point samples; keep them on wet ice. 2. Add an appropriate antibody (see Note 14) to immunoprecipitate the remaining target protein; incubate on ice for 1 h. 3. Add sepharose-protein A or -protein G, as appropriate for your antibody; incubate on ice with occasional shaking or on a rotating wheel at 4°C for 30 min. 4. Wash samples twice with 1 mL of E1A buffer (see Note 15). 5. Remove as much E1A buffer as possible and add 20 µL of gel-loading buffer. 6. Run samples and loading controls on SDS-PAGE (at a polyacrylamide percentage suited to the target protein). 7. Dry gel and expose on film.

4. Notes 1. It used to be necessary to first transcribe your DNA, to purify the resulting mRNA, and then to translate your protein of interest from the purified mRNA. Although it is perfectly feasible to do this if necessary, the TNT kits provide a number of advantages that should be considered. Although they are expensive, they save a considerable amount of time. In addition, working with purified RNA is notoriously tricky and requires very good technical ability to keep batches consistent and free from RNases. If you use a kit, you do not have to develop the same level of technical expertise before you can get proteins to use in your assays. 2. Although in vitro translated (ivt) proteins can be labeled with 14C-leucine if they have neither methionine nor cysteine residues, it is generally more effective to label them with 35S. Assessment of the quality of translation is quicker, and thus your experiments can start sooner, while your translation product is still fresh. A radionuclide specifically intended for in vitro translation is advisable, such as the

Substrate Degradation Induced by HPV E6 Oncoproteins 35S-methionine/cysteine

TransLabel sold by ICN or the

35S-cysteine

415

sold by Amersham. 3. There are many vectors available for the in vitro translation of proteins, but the plasmids of choice are pSP64- or pCDNA3-based plasmids containing your genes of interest cloned downstream of the SP6 or T7 promoters, respectively. In general, all else being equal, SP6 promoters translate slightly better than T7 promoters in rabbit reticulocyte lysate and markedly better in wheat germ extract, but the pCDNA3 plasmids have the advantage of containing the cytomegalovirus (CMV) promoter, allowing the same plasmid to be used for in vivo expression in eukaryotic cells. In our hands, the T3 promoter is weaker and rarely produces equivalent levels of protein. In any case, it is worth ensuring that your gene of interest contains a Kozak sequence with the initiation codon (4–6), as this can increase translation levels considerably. 4. Alkaline SDS lysis and cesium chloride gradient centrifugation are the best methods of preparing clean and concentrated DNA for in vitro transcription/translation (see Chapter 27 for protocol). In the case of pCDNA3-based plasmids, this has the additional advantage of producing DNA suitable for transfecting into most eukaryotic cell lines. If you are constructing a vector for protein expression, you will almost certainly mini-prep your DNA for restriction enzyme digestion to ensure that your DNA fragment is present. Mini-prep method: spin out 1.5 mL of an overnight bacterial culture (2 min at full speed in a bench-top microfuge). Resuspend the pellet in 100 µL solution I (50 mM glucose, 25 mM Tris HCl (pH 8.0), 10 mM EDTA); after 5 min, add 200 µL of solution II (0.2 M NaOH, 1% SDS) on ice; after a further 5 min, add 150 µL of solution III (5 M potassium acetate, pH 4.8). Mix by shaking and spin in a bench-top microfuge for 5 min. Transfer 400 µL of supernatant into 1 mL ethanol and freeze on dry ice for 5 min. Spin in microfuge for 10 min; resuspend the pellet in 20 µL water. This DNA prep can also be used to check whether the protein is translated from any of your clones (this is particularly useful if you are cloning the fragment into a single site). However, the DNA should be treated for 1 h with RNase and then subjected to phenol/chloroform extraction and ethanol precipitation, before the translation reaction, which should, incidentally, contain at least twice the usual amount of RNase inhibitor. Remember to include a DNA construct that is known to be translatable in vitro as a positive control. You should also be aware that although a positive result is very useful (you only have to maxiprep DNA that is known to work), a negative result means nothing—you may just not be a very good miniprepper! So use this as a short-cut only if you are sure of yourself and can spare the reagents to do it. 5. Concentrated DNA stocks (either in sterile TE buffer or in sterile water) should be stored at –20°C, and more dilute working stocks (1 µg/µL in water) should be prepared from them, and also kept at –20°C. In general, it is advisable to keep the working stock in small aliquots to reduce the number of freeze–thaw cycles. Excessive freeze/thawing of the DNA, especially of the more dilute working stock, can damage it and reduce the level of translated product.

416

Thomas and Banks

6. For simple degradation assays, involving only one target protein plus an E6 protein, it is usual to use the amino acid most abundant in that protein: HPV E6 proteins are always translated using 35S-cysteine. However, if you intend to compare several target proteins, it is advisable to label each as equally as possible. For example, MAGIs-1, -2, and -3 have 13, 14, and 14 Cys residues, respectively, compared with 15, 25, and 22 Met residues. Cys labeling is thus the choice if comparing MAGI-1 with the others, whereas Met labeling might be preferable when looking at the degradation of MAGI-2 or MAGI-3 alone. 7. Always use fresh lysate. Although proteins can be translated in reticulocyte lysate that has been thawed more than once, you are unlikely to get good degradation from such translates. Other components of the proteasome pathway found in the lysate are very susceptible to freezing and thawing. Equally, although freezing once at –80°C after translating your proteins should not affect things too much, do not re-use previously thawed and refrozen aliquots of protein. Use them instead in GST-pulldown assays, or as positive controls or molecular-weight markers for Western blots. 8. Always set up at least one translation reaction primed with water, rather than DNA. Water-primed lysate (wpl) is used to equalize the volumes of degradation assays. Experience alone will tell you how much you are likely to need, so it is worth freezing it in several aliquots. 9. The standard size of one translation reaction as recommended in the Promega kits is 50 µL; however, half of that is often more than enough, particularly for target proteins. To check the levels of protein expression, take 1 µL of the translation reaction and run it on an SDS-PAGE gel, dry the gel, and assess your translation efficiency by exposure overnight on film, or by phosphorimager analysis. The latter has the advantage of being quicker and more quantifiable; a film, however, gives a clearer image, thus allowing you to assess the quality, as well as quantity, of your translated protein. 10. Certain reports have suggested that extended or even overnight incubations are required to see degradation, even of p53 in the presence of HPV-16 E6. By our standards, these count as failed assays: I would seriously doubt whether any E6-induced degradation of target protein could be reliably distinguished from the “noise” of simple protein instability under such conditions. If you do not see any degradation of a known target protein in 2 h, then the conditions of your assay probably need to be optimized. An indication of the approximate half-lives of various HPV E6 targets in in vitro degradation assays is shown in Fig. 1. 11. Because you are assaying degradation, all samples should be kept on ice at all times when not actually undergoing the degradation incubation at 30°C! 12. A number of early publications gave in vitro degradation methods using various buffered solutions to dilute the translated proteins and reduce the supposed inhibitory effects of hemoglobin upon in vitro degradation. In our hands, hemoglobin has never been inhibitory, but the use of these buffers has completely blocked any degradation. The notion may have arisen from the fact that running large volumes of radiolabeled reticulocyte lysate on an SDS-PAGE gel results in


417

a smear that makes it very difficult to see any individual protein bands. We have addressed this point by using an immunoprecipitation step to remove excess hemoglobin. 13. With degradation assays having long incubations, such as Dlg, or MUPP1 assays, it is advisable to spin the reaction tubes briefly in a bench-top microfuge before taking samples, as moisture evaporates from the reaction mixture and may condense on the lid, thus altering the concentration of the reaction mixture. In shorter assays, with targets such as MAGI-1, the variation in concentration is negligible in comparison with the variation in time points that would be caused by doing this. 14. The choice of antibody is important, as it is obviously necessary to immunoprecipitate as much of the protein as possible. Often a high-affinity polyclonal serum is very effective; alternatively, a mixture of monoclonal antibodies recognizing different epitopes on the protein is used. For example, in p53 degradation assays, we use either a polyclonal rabbit serum or a mixture of the pAb1801, pAb1802, and pAb1803 monoclonal antibodies. 15. Washing of these immunoprecipitations should not be done with the same rigor as would be used in a true immunoprecipitation reaction; two washes of 1 mL E1A buffer are usually adequate to remove excess reticulocyte lysate without losing significant amounts of the target protein.

References 1. Hershko, A. (1997). Roles of ubiquitin-mediated proteolysis in cell cycle control. Curr. Opin. Cell Biol. 9, 788–799. 2. Obaya, A. J. and Sedivy, J. M. (2002). Regulation of cyclin-Cdk activity in mammalian cells. Cell. Mol. Life Sci. 59, 126–142. 3. Banks, L., Pim, D., and Thomas, M. (2003). Viruses and the 26S proteasome: hacking into destruction. Trends Biochem. Sci. 28, 452–459. 4. Kozack, M. (1987). An analysis of 5'-noncoding sequences from 699 vertebrate messenger RNAs. Nuc. Acids Res. 15, 8125–8148. 5. Kozack, M. (1990). Downstream secondary structure facilitates recognition of initiator codons by eukaryotic ribosomes. Proc. Natl. Acad. Sci. USA 87, 8301–8305. 6. Kozack, M. (1991). An analysis of vertebrate mRNA sequences: intimations of translational control. J. Cell. Biol. 115, 887–903. 7. Thomas, M., Laura, R., Hepner, K., et al. (2002) Oncogenic human papillomavirus E6 proteins target the MAGI-2 and MAGI-3 proteins for degradation. Oncogene 21, 5088–5096. 8. Pim, D., Thomas, M., Javier, R., Gardiol, D., and Banks, L. (2000). HPV E6 targeted degradation of the discs large protein: evidence for the involvement of a novel ubiquitin ligase. Oncogene 19, 719–725. 9. Lee, S. S., Glaunsinger, B., Mantovani, F., Banks, L., and Javier, R. T. (2000) Multi-PDZ domain protein MUPP1 is a cellular target for both adenovirus E4-ORF1 and high-risk papillomavirus type 18 E6 oncoproteins. J. Virol. 74, 9680–9693.

Induction or Inhibition of Apoptosis by HPV Proteins

419

30 Measuring the Induction or Inhibition of Apoptosis by HPV Proteins Anna M. Kowalczyk, Geraldine E. Roeder, Katie Green, David J. Stephens, and Kevin Gaston Summary Many viral proteins influence the cellular pathways that control cell proliferation and cell death. Some viral proteins trigger apoptotic cell death, and this may be important in host defense and viral spread. In other cases, viral proteins inhibit apoptosis. In this chapter, we will describe some of the methods that can be used to investigate the induction or inhibition of apoptosis by papillomavirus proteins.

1. Introduction Viral replication often requires host cell proliferation, and viruses have evolved multiple strategies to induce cell-cycle progression. Viral infection can also trigger cell suicide, and many viruses encode proteins that inhibit cell death. However, in some cases viruses might induce cell death in order to promote the spread of viral particles or viral genomes, as uninfected cells take up the remains of their dead neighbors (1). Cell death can occur via either of two pathways: 1. Necrosis occurs after severe or sudden injury, such as physical or chemical trauma, and is characterized by cell swelling and rupture followed by release of the cell contents and the stimulation of an inflammatory response (2). 2. Apoptosis is a more controlled form of cell death and is characterized by cell shrinkage, membrane boiling or blebbing, as well as nuclear condensation and DNA fragmentation (2,3). The endpoint of apoptosis is the formation of membrane-bound apoptotic bodies that are rapidly engulfed by phagocytes.

This chapter describes some of the methods that can be used to study the induction or inhibition of apoptosis by viral proteins. Several studies have From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

419

420

Kowalczyk et al.

shown that under some circumstances, the human papillomavirus (HPV) E2 and E7 proteins can induce apoptosis, whereas the HPV E6 protein can inhibit apoptosis (reviewed in ref. 4). Although these experiments typically involve protein over-expression and they are thus not necessarily representative of events in the HPV life cycle, apoptosis is an easily measurable endpoint, which allows a quantitative assessment of the effects of these proteins on the cell (5,6). Techniques such as flow cytometry examine an entire population of cells and can determine the percentage of cells that are undergoing apoptotic cell death (7). In contrast, the death of individual cells can be observed using microscopy, either in combination with specific markers or simply by observing changes in cell morphology (5). Alternatively, the activity of apoptosisassociated enzymes such as caspases can be determined. We describe several of these different assays and outline their relative strengths and weaknesses. 2. Materials 1. Green fluorescent protein (GFP) and red fluorescent protein (RFP) expression plasmids (e.g., pEGFP-C1 and pHcRED-C1, Clontech). 2. 6-Well plates. 3. 100-mm-Diameter tissue-culture dishes. 4. Sterile 75-cm2 tissue-culture flasks. 5. 22 × 22 mm Glass cover slips. 6. Mowiol adhesive: 2.4 g Mowiol (Calbiochem), 6 g glycerol, 6 mL H2O, 12 mL 0.2 M Tris-HCl (pH 8.5). 7. 1-mm-Thick glass microscope slides. 8. Phosphate-buffered saline (PBS): 137 mM NaCl, 2.7 mM KCl, 1.5 mM KH2PO4, 8.1 mM Na2HPO4 (pH 7.4). 9. 4% w/v paraformaldehyde in PBS. 10. Bisbenzimide (Hoechst 33258) (Sigma) (keep in darkness). 11. Dead End™ Fluorometric terminal uridyl-nucleotide end labeling (TUNEL) System: (9.6 mL equilibration buffer, 300 µL nucleotide mix (6 × 50 µL), 1500 U terminal deoxynucleotidyl transferase (TdT) enzyme (3 × 500 U), 70 mL 20X SSC, 10 mg proteinase K, plastic cover slips) (Promega). 12. 4% Methanol-free formaldehyde solution in PBS (pH 7.4). 13. 0.2% Triton® X-100 solution in PBS. 14. 0.1% Sodium borohydride in PBS. 15. 2X SSC: 0.33M NaCl, 0.03 M tri-sodium citrate. 16. 0.1% Triton® X-100, 5 mg/mL bovine serum albumin (BSA) in PBS. 17. Propidium iodide (Sigma) (50 µg/mL in PBS). 18. H2B-GFP HeLa cells (8). 19. Glass-bottom dishes (MatTek). 20. Borosilicate glass capillaries or Femtotips (Eppendorf). 21. Capillary puller (optimally, a programmable horizontal Flaming-Brown type such as Sutter P-97).


421

22. High-vacuum silicone grease. 23. Pure plasmid DNA (Qiagen midi/maxi kit or cesium chloride preparation). 24. Microinjection and micromanipulation devices: Eppendorf Femtojet, Eppendorf micromanipulator 5171 and controller, Eppendorf Injectman. 25. Phase-contrast microscope for microinjection and fluorescence microscope for live-cell imaging, equipped with fluorescein isothiocyanate (FITC), tetramethylrhodamine isothiocyanate (TRITC), and Cy3 filter sets. 26. Microinjection and imaging medium: phenol red-free MEM (Sigma), 30 mM HEPES (pH 7.4), 0.5 g/L sodium bicarbonate. For longer term imaging, this is supplemented with fetal calf serum (10%). 27. Microloader tips (Eppendorf). 28. Diamond-tipped pen (Fischer). 29. Trypsin/ethylenediamine tetraacetic acid (EDTA) (1%). 30. BD ApoAlert™ Caspase Colorimetric Assay Kit (100 mL cell lysis buffer, 16 mL 2X reaction buffer, 800 µL DTT (1 M), 2 × 100 mL dilution buffer, 1 mL caspase8 substrate, IETD-pNA) (BD Biosciences). 31. Cell scrapers. 32. 100-µL Quartz cuvet. 33. Spectrophotometer. 34. FACScalibur (Becton Dickinson). 35. Fluorescence activated cell sorter (FACS) tubes (Falcon). 36. Leica Q550 fluorescent microscope with a Leica DC500 camera. 37. Ethanol (70%). 38. Methanol.

3. Methods 3.1. Cell Morphology Apoptosis is characterized by a number of morphological events, including chromosomal DNA condensation, DNA fragmentation, cell membrane blebbing, cell shrinkage, and the formation of small apoptotic bodies (3). Examining cell morphology is therefore a useful method for the identification of cells undergoing apoptosis ([5,7,9] and see Note 1).

3.1.1. Fluorescence Microscopy Fluorescence microscopy allows the identification of apoptotic cells on the basis of their morphology. In addition to any plasmids expressing HPV proteins, cells can be co-transfected with plasmids expressing green or red fluorescent protein (GFP and RFP, respectively) (see Note 2). GFP and RFP are expressed uniformly across the cell and aid observation of the cytoplasm when cells are viewed under FITC or TRITC filter sets, respectively. Cells should be fixed and stained with bisbenzimide (see Note 3). Bisbenzimide stains cell nuclei, allowing the visualization of nuclear morphology under ultraviolet (UV) illumination and hence assessment of the apoptotic status of cells (Fig. 1).

422

Kowalczyk et al.

Fig. 1. Chromatin condensation and membrane blebbing in apoptotic cells. HeLa cells were transiently co-transfected with pEGFP-C1 (Clontech) and pWEB-E2 (5) using Fugene 6. Twenty-four hours later, the cells were fixed, stained with bisbenzimide stain, and mounted on slides. Cells were visualized using a ×40 oilimmersion lens fitted to an epifluorescent confocal microscope. (A) Bright field microscopy. (B) The same field of cells visualized using a fluorescein isothiocyanate (FITC) filter set. The transfected cell expressing green fluorescent protein (GFP) is clearly visible. (C) The same field of cells stained with bisbenzimide and visualized using a 4',6-diamidino-2-phenylindole hydrochloride (DAPI) filter set. N indicates a normal nucleus. “A” indicates an apoptotic nucleus showing chromatin condensation. 1. Seed 3 × 105 cells onto 22 × 22 mm cover slips in six-well plates and incubate overnight at 37°C in 5% CO2. 2. Following the induction of protein expression (see Note 4) and an appropriate incubation period, discard the medium by inverting the plate. 3. Wash the cells twice with PBS (see Note 5). 4. Fix the cells with 1 mL 4% w/v paraformaldehyde/PBS per well (see Note 5). 5. Incubate at 22°C for 30 min. Meanwhile, defrost bisbenzimide in darkness. 6. Wash the cells twice with PBS (see Note 5). 7. To each well (pipetting onto well side) add 1 mL bisbenzimide (1 µg/mL) and incubate in darkness at 22°C for 30 min (see Note 3). 8. Wash the cells three times in PBS, but to facilitate removal of the cover slip, do not discard the third wash. 9. Using a p1000 pipet, apply a 10-mm-diameter drop of Mowiol adhesive to a glass slide (see Note 6). 10. While ensuring that the cover slip is still covered in PBS, tilt the plate slightly to one side and, using a pair of very finely tipped forceps, gently lift the edge of the cover slip. 11. Remove the cover slip from the well, blotting the edge onto a piece of tissue to remove excess moisture. Invert the cover slip and slowly lower onto Mowiol, allowing one edge to stick before using forceps to carefully lower the rest. Gently press the cover slip onto Mowiol with forceps to remove any air bubbles. 12. Leave in the dark at 22°C overnight.


423

Fig. 2. Fluorometric terminal uridyl-nucleotide end labeling (TUNEL) assays. HeLa cells were transiently co-transfected with pHcRED C1 and pWEB-E2 (5) using Fugene 6. Eighteen hours later, the cells were fixed, labeled with fluorescein-12-dUTP, stained with bisbenzimide, and mounted on slides. Cells were visualized using a ×40 lens fitted to an epifluorescent microscope. (A) A fluorescein-12-dUTP-labeled apoptotic cell visualized using a fluorescein isothiocyanate (FITC) filter set. (B) The same field of cells stained with bisbenzimide and visualized using a 4',6-diamidino-2phenylindole hydrochloride (DAPI) filter set. (C) The same field of cells visualized using a tetramethylrhodamine isothiocyanate (TRITC) filter set. A transfected cell expressing red fluorescent protein (RFP) is clearly visible. “N” indicates a normal nucleus. “A” indicates an apoptotic/TUNEL-positive nucleus.

13. Before examining, wipe over the slide with 70% ethanol to remove any excess Mowiol. 14. Count apoptotic cells (see Note 7).

3.1.2. Fluorometric TUNEL Assays TUNEL assays use DNA fragmentation as a marker of apoptosis (Fig. 2 and see Note 8). 1. Seed 3 × 105 cells onto 22 × 22 mm glass cover slips in a six-well plate and incubate overnight at 37°C in 5% CO2. 2. Following the induction of protein expression (see Note 4) and an appropriate incubation period, discard the medium by inverting the plate. 3. Wash the cells twice with PBS (see Note 5). 4. Fix the cells by adding 1 mL 4% w/v methanol-free formaldehyde solution in PBS (pH 7.4) and incubating at 4°C for 30 min. 5. Wash twice with PBS for 5 min (see Note 5). 6. Permeabilize the cells by adding 0.2% Triton® X-100 solution in PBS and leaving on ice for 10 min. 7. Wash twice in PBS (see Note 5). 8. Block for 10 min in 0.1% sodium borohydride in PBS. 9. Wash twice in PBS (see Note 5).

424

Kowalczyk et al.

10. Pipet off excess PBS and cover the cells with 100 µL of equilibration buffer (see Note 9). 11. Incubate at 22°C for 10 min. 12. Thaw the nucleotide mix and prepare sufficient TdT incubation buffer (see Note 9). 13. After 10 min, remove the equilibration buffer (see Note 10) and add 50 µL of TdT incubation buffer (see Note 9). 14. Wrap the plate in foil to protect the cells from light, and incubate at 37°C for 60 min to allow the tailing reaction to occur. 15. Add 2 mL 2X SSC to each well. 16. Wash the cells three times in 0.1% Triton® X-100, 5 mg/mL BSA in PBS. 17. Wash the cells twice in PBS (see Note 5). 18. Process the cover slips as described in steps 9–14 in the previous method.

3.1.3. Microinjection of Single Cells for Live-Cell Imaging Capillary microinjection of single living cells is a technique in which DNA or proteins are injected using glass capillaries with very fine tip diameters ([10,11] and see Note 11). We use the example here of imaging changes in cell shape and chromatin structure following microinjection of the HPV-16 E2 protein (Fig. 3). 1. Equilibrate the microscope to 37°C. 2. Grow H2B-GFP HeLa cells (see Note 12) on glass-bottom dishes at 37°C in 5% CO2 to approx 70% confluence (see Note 13). 3. Pull capillaries to the appropriate size. For the Sutter P-97, use a heat setting 5°C above the ramp temperature for the filament. Alternatively, use Eppendorf Femtotips (see Note 11). 4. Dilute plasmid DNA (encoding E2 and HcRed) (see Note 14) to 50 µg/mL each in ddH2O and centrifuge at 25,000g for 30 min; transfer the supernatant to a fresh tube. 5. Load a capillary with 2 µL of diluted plasmid DNA using a microloader pipet tip (see Notes 15–17). 6. Fix the capillary to the micromanipulator (see Note 18). 7. Transfer cells to imaging medium (about 4 mL in a dish is needed) and mark a cross on the base of the dish using a diamond-tipped pen for location of injected cells. 8. Using a ×10 phase-contrast objective, locate the capillary above the cells (see Note 19) and inject cells directly into the nucleus. 50–100 cells should be sufficient per dish. 9. Return cells to growth medium and incubate at 37°C and 5% CO2 for 2 h. Further dishes of cells can now be used. 10. Transfer the cells to pre-warmed imaging medium. Completely fill the dish and seal the lid back on, using high-vacuum silicone grease (see Note 20). 11. Using a ×100 1.4 N.A. oil-immersion lens, locate injected cells by focusing initially with phase contrast to locate the cross on the cover slip, followed by HcRed fluorescence to identify injected cells.


425

Fig. 3. Time-lapse video microscopy of apoptosis in microinjected cells. Still pictures taken every 10 min were extracted from a larger data set of pictures taken every 30 s. Two HeLa cell nuclei are present in each frame. The lower of the two nuclei was microinjected with a plasmid expressing the human papillomavirus (HPV)-16 E2 protein (pWEB-E2). (A) H2B-green fluorescent protein (GFP) visualized using a fluorescein isothiocyanate (FITC) filter set. Chromatin condensation is clearly visible at 12 h 40 min and increases with time. (B) Ds-Red visualized using a tetramethylrhodamine isothiocyanate (TRITC) filter set. Blebbing of the plasma membrane is first apparent at 13 h 30 min and is clearly visible at later time points. 12. Once suitable cells have been found, start time-lapse imaging of both GFP and HcRed fluorescence (see Note 21). Images should be acquired with very short exposure times (e.g., 50 ms) with a delay between frames of 30 s (120 frames/h). Continue imaging through apoptosis. This may take up to 5 h (600 frames) (see Notes 22 and 23). 13. Individual time-lapse sequences of each channel can then be merged to an RGB image and played back (Fig. 3).

3.2. Flow Cytometry Flow cytometry is a technique in which individual cells in a suspension can be analyzed (7). To measure apoptosis within a population of cells, propidium iodide is used to stain the DNA (Fig. 4) (see Note 24). 1. Grow 3 × 106 cells in a 75-cm2 flask at 37°C and 5% CO2 until 70% confluent. 2. Following the induction of protein expression (see Note 4) and an appropriate incubation period, remove and keep the medium (see Note 25). 3. Wash the cells twice with 1X PBS (see Note 5). 4. Trypsinize the cells and add to the medium collected in step 2.

426

Kowalczyk et al.

Fig. 4. Flow cytometry. 2.3 × 106 HeLa cells were transiently transfected with (A) 7.5 µg of pWEB or (B) 7.5 µg of pWEB-E2 using Fugene 6. Twenty-four hours posttransfection, the cells were trypsinized and harvested from the medium, then fixed and stained with propidium iodide before analysis by flow cytometry. 5. Pellet the cells by centrifugation at 1500g for 5 min at 20°C. 6. Discard the supernatant and wash the pellet in PBS, then repeat step 5. 7. Fix cells by resuspending in 1 mL ice-cold methanol and incubating at –20°C for 5 min. 8. Pellet the cells by centrifuging at 3000g for 10 min at 4°C. 9. Resuspend the cell pellet in 3 mL propidium iodide (50 µg/mL in PBS) and leave at 4°C for 30 min. 10. Harvest the cells by centrifuging at 3000g for 10 min at 4°C. 11. Discard the supernatant and resuspend the cell pellet in 300 µL PBS. 12. Transfer to a fluorescence activated cell sorter (FACS) tube. 13. Keep in the dark until analysis by flow cytometry.

3.3. Caspase Assays Caspases are a family of cysteine proteases that, upon activation, cleave cellular substrates after an aspartic acid residue (12). Caspases are synthesised as inactive zymogens and are activated in a number of apoptotic pathways. Caspase activity is relatively easy to assay and is therefore a useful indicator of apoptosis (Fig. 5) (see Note 26). 1. Seed 2 × 106 cells in a 100-mm-diameter dish and incubated overnight at 37°C in 5% CO2. 2. Following the induction of protein expression (see Note 4) and an appropriate incubation period, discard the medium and wash the cells twice in PBS (see Note 5). Following washing, leave the dishes upside down on a piece of absorbent paper to remove any remaining PBS. 3. To each dish add 50 µL of ice-cold lysis buffer (see Note 27).


427

Fig. 5 Caspase assays HeLa cells were treated with the concentrations of staurosporine (a compound known to induce apoptosis) shown for 4 h and then assayed for caspase-8 activity as described in the text. 4. 5. 6. 7. 8.

Scrape off the cells using cell scraper and collect the lysate in a 1.5-mL tube. Incubate on ice for 10 min (see Note 28). Centrifuge the lysate at 13,000g for 10 min at 4°C. Transfer the supernatant to a fresh tube and place on ice. To each supernatant add 50 µL dithiothreitol (DTT)/reaction buffer mix (see Note 27). 9. Add 5 µL 4 mM caspase substrate (see Note 27). 10. Incubate at 37°C for 1.5 h. 11. Read samples at 405 nm in a 100-µL quartz cuvet using a spectrophotometer.

4. Notes 1. Cells should be seeded on glass cover slips before the introduction of HPV proteins by transient transfection, viral transformation, microinjection, or other protein/DNA delivery method. Our research typically uses plasmids to express HPV proteins. However, we have also used adenoviral vectors and the direct delivery of purified proteins. Apoptosis is typically evident 8–12 h after protein expression, and typically reaches a maximum after 12–48 h. 2. pEGFP-C1 (Clontech) and pHcRED-C1 (Clontech) are suitable GFP and RFP expression plasmids, respectively. Care should be taken not to express excessive amounts of GFP or RFP, since this can induce apoptosis in some cell types. We typically transfect cells with 300 ng of plasmid DNA. GFP/RFP expression is also indicative of successful transfection. 3. Alternatively, 4',6-diamidino-2-phenylindole hydrochloride (DAPI) can be used. 4. HPV protein expression can be achieved by several means, including transient or stable transfection, microinjection of plasmid DNA, viral infection, or the direct

428

5. 6. 7.

8.

9. 10. 11.

Kowalczyk et al. delivery of proteins using microinjection or commercially available carriers such as BioPORTER® Protein Delivery Reagent (Gene Therapy Systems, Inc.). Typically, transient transfection can be achieved using reagents such as Fugene (Roche), Tfx (Promega), Transfast (Promega), Metafectene (Biontex), Geneporter (Gene Therapy Systems, Inc), or Lipofectin (Invitrogen). To avoid dislodging the cells, wash the cover slips by tilting the plate and gently applying solutions to the side of the well. Alternatively, Vector Shield can be used and cover slip edges sealed with nail varnish to prevent movement of cover slip. To quantify apoptosis induced or inhibited by HPV proteins, at least 100 transfected cells (those expressing GFP/RFP) should be counted and their cytoplasmic and nuclear morphology assessed for apoptotic characteristics (Fig. 1). A percentage value can then be obtained for the number of transfected cells that are apoptotic. This should be repeated at least twice per slide and on at least two independent slides. Untransfected cells and cells transfected with empty vectors (or infected with control viruses) should be counted to determine the background level of apoptosis in any particular cell line. In the Dead End Fluorometric TUNEL System (Promega), the enzyme terminal deoxynucleotidyl transferase (TdT) catalytically attaches fluorescein-12-dUTP to the 3' OH group of DNA fragments. Fluorescein-12-dUTP-labeled DNA can then be visualized by fluorescence microscopy, using a FITC filter set (Fig. 2A). As a comparison, propidium iodide or bisbenzimide can be used to stain all cells, which can then be visualized by fluorescence microscopy, using TRITC or DAPI filter sets (Fig. 2B). When looking at HPV protein–induced apoptosis, cells can be co-transfected with a plasmid expressing RFP to allow identification of the transfected cells by fluorescence microscopy, using a TRITC filter set (Fig. 2C). These reagents are supplied in the Dead End™ Fluorometric TUNEL System (Promega). Tilt the plate and remove the buffer using a tissue. Do not allow cells to dry out. Microinjection is now a widely available technique, with many suppliers providing solutions for microinjection and micromanipulation. We use a system from Eppendorf (Hamburg, Germany), but others are available from suppliers such as BioRad or Narishige. Prepulled capillaries are available from suppliers (e.g., Femtotip from Eppendorf) or can be pulled from borosilicate glass capillaries (internal diameter 0.94 mm) using a capillary puller, such as the Sutter Instruments P-97 or Narishige PN-30. Best results are obtained with programmable horizontal pullers, but vertical and manual options are available. We use capillaries that contain a central filament, which facilitates backfilling of the sample. Pre-pulled capillaries offer excellent reproducibility but are expensive. The benefit of pulling one’s own capillaries is the ability to modify the tip size and shape by adjusting the settings used for pulling. Settings on Flaming-Browntype pullers (such as the Sutter) can be changed to produce capillaries of the desired form. We commonly find that a single setting will produce capillaries for microinjection of both DNA and protein into cultured mammalian cells.


429

12. For the identification of chromatin structure, we have used a stable HeLa cell line expressing a histone H2B-GFP fusion protein (H2B-GFP) (8). In order to view the overall shape of the cells, we co-express HcRed (Clontech, Palo Alto, CA), which fills the cytoplasm. This also facilitates the identification of injected cells. 13. A key consideration when performing microinjection experiments is the nature of the substrate for cell growth. Microinjection of suspension cultures is possible but obviously more difficult than using adherent cells. Adherent cells can be grown on glass cover slips before microinjection. In order to image through a plastic dish and glass cover slip, one must use a long-working-distance objective. This is usually not a problem, since low power (×10) phase-contrast objectives allow the user to inject many cells without needing to change the position of the microscope stage. However, an alternative is to use specially designed dishes that have a glass cover slip instead of a plastic base, allowing use with highnumerical-aperture (N.A.) oil- or water-immersion lenses. We routinely grow cells on glass-bottom dishes (MatTek) rather than cover slips, and we have developed protocols for long-term imaging of these cells. These dishes are available from a number of suppliers, including MatTek (Ashland, MA) and World Precision Instruments (Sarasota, FL), and can be coated with collagen or polylysine if required. 14. An alternative to the use of genetically encoded reporters is to use a non-toxic reporter of injection, such as fluorescently labeled dextran or serum albumin. 15. We routinely load 2 µL into a capillary using a gel loader tip and inject plasmid DNA at capillary concentrations of 20–50 µg/mL and protein solutions at 1 mg/mL. Prior to microinjection, cells are transferred to buffered medium, pre-equilibrated to the temperature to be used for injection. Optimally, cells should be removed from the incubator for only 10–15 min for injection, during which time sufficient cells can be injected for most experiments. There is a trade-off here between the number of cells that can be injected and the time that cells are left in microinjection medium. 16. Injection of plasmid DNA requires a lag time for expression of E2. Expression from microinjection tends to occur relatively quickly; GFP fluorescence can be detected approx 30 min after injection of plasmid DNA. The major caveat is that, unless inhibitors of protein synthesis are added, expression levels continue to rise as the experiment progresses. This then does not provide an accurate measure of whether a critical concentration of E2 is needed for the induction of apoptosis or whether it is solely a time-dependent effect. The advantage of DNA injection over protein injection is that it is simpler in terms of both sample preparation and also injection itself (protein solutions tend to be more viscous). 17. Plasmids should be of high purity for microinjection. Cesium-chloride-purified plasmid DNA works well, as do plasmids purified using kits such as Qiagen Maxi and Midi kits, as well as those from other suppliers. Plasmids purified with these kits should be centrifuged at 20,000g after resuspension to remove any impurities that pass through the columns during use. 18. The manipulators are mounted (either left- or right-handed, according to user preference and microscope configuration) on a suitable microscope frame. Most

430

19.

20.

21.

22.

23.

Kowalczyk et al. micromanipulators are adaptable to a range of microscope frames. The technique of microinjection relies heavily on hands-on instruction and practice. Many of the specific details of methods need to be determined empirically. Location of the capillary once mounted on the system is often difficult. Position the capillary above the lens aperture and lower to a point above the cells. Move the capillary in large sweeps across the field of view; a shadow of the out-offocus needle will be visible. Center this shadow and slowly lower the capillary until nearly in focus. The z level for the capillary can be set by gently lowering onto a cell adjacent to the nucleus; a bright spot should be visible before the cell is punctured. Set this as your z-limit. Move the capillary away and try to inject a nearby cell. Pressure settings and minor adjustments of the z-limit may be needed to achieve optimal injection. Good injection will cause a visible but not too dramatic swelling of cells. Use the following Injectman settings: compensation pressure, 50 hPa; injection pressure, 100–200 hPa; injection time, 0.1–0.4 s. It is imperative to have controlled environmental conditions for live cell imaging. Our microscope is housed in a heated Perspex box (Solent Scientific, Portsmouth, UK). Dishes are filled completely with imaging medium (supplemented with fetal calf serum) and sealed. This prevents gas exchange, eliminating the need for control of CO2 concentration, and maintains HeLa cells in a continually dividing state for periods of >48 h. The key problem in these experiments is focal stability. Long-term imaging requires stable focus, and continual switching on and off of temperature and CO2 regulators can cause significant focal drift. Imaging systems should be switched on several hours before imaging commences to ensure thorough equilibration of temperature. The use of spectrally distinct fluorophores (GFP and HcRed) and sequential acquisition enables specific analysis of each in isolation. Combination of these images into an RGB image facilitates accurate identification of the temporal sequence of events. For live cell imaging we use a TILL Photonics (Gräfelfing, Germany) imaging system based on an Olympus IX-70 frame with monochromatic illumination (eliminating the need for excitation filters) and a double dichroic emission filter tailored for FITC and Cy3 (Chroma 5100; Chroma, Rockingham, VT). However, the specific experiment and availability of equipment will largely determine which imaging system is used. In this example, we employ a wide-field microscope, but scanning or spinning-disk confocal microscopes could also be used. Our system provides relatively low levels of excitation light through a fiber-optic coupling from the monochromator. This facilitates long-term imaging by reducing photodamage. Alternatives are to use appropriate neutral-density filters in the illumination path or greatly reduce exposure times. When imaging, alternative lenses are suitable but a ×60 or ×100 high-N.A. lens will provide suitable resolution and light throughput. Exposure times of 50 ms, combined with appropriate grayscaling of images for presentation gives good images while minimizing light exposure. Individual settings should be determined for each experiment. Delays in the time-lapse sequence will need to be included to accommodate the timeframe of imaging.


24.

25. 26.

27. 28.

431

Thirty-second or 1-min delays between acquisitions will reduce the amount of data acquired. One must ensure that the time delay is big enough to limit the data but acquisition is rapid enough to ensure that one has adequate time resolution. Propidium iodide binds stoichiometrically to DNA, providing a quantitative measurement of the DNA content within a cell, a key indicator of cell-cycle progression and of apoptosis. Cells that are not progressing through the cell cycle are said to be in G0. When cell division is triggered, the cell enters the G1 phase of the cell cycle and prepares for DNA replication. Upon completion of G1, the cell enters S phase, where the DNA content of the cell is doubled before the cell enters G2. Finally, the cell enters mitosis (or M) and divides. Cells undergoing apoptosis have a DNA content less than that seen in G0/G1 cells and are therefore referred to as the sub-G0 population (see Fig. 4). Late-stage apoptotic cells detach from the plastic and float in the growth medium. We use the BD ApoAlert Caspase Colorimetric Assay (BD Biosciences), in which spectrophotometric detection of the chromophore p-nitroaniline (pNA), produced after the cleavage of a specific substrate, is used to quantify caspase activity (Fig. 5). However, several similar assays are commercially available. These reagents are supplied with the BD ApoAlert™ Caspase Colorimetric Assay Kit (BD Biosciences). Meanwhile, defrost the substrate and make your 1:100 dilution of the DTT/reaction buffer mix.

References 1. Roulston, A., Marcellus, R. C., and Branton, P. E. (1999) Viruses and apoptosis. Annu. Rev. Microbiol. 53, 577–628. 2. Searle, J., Kerr, J. F. R., and Bishop, C. J. (1982) Necrosis and apoptosis—distinct modes of cell-death with fundamentally different significance. Pathol. Annu. 17, 229–259. 3. Earnshaw, W.C. (1995) Nuclear changes in apoptosis. Curr. Opin. Cell Biol. 7, 337 43. 4. Dell, G. and Gaston, K. (2001) Human papillomaviruses and their role in cervical cancer. Cell. Mol. Life Sci. 58, 1923–1942. 5. Webster, K. J., Parish, J., Pandya, M., Stern, P. L., Clarke, A. R., and Gaston, K. (2000) The human papillomavirus (HPV) 16 E2 protein induces apoptosis in the absence of other HPV proteins and via a p53-dependent pathway. J. Biol. Chem. 275, 87–94. 6. Desaintes, C., Demeret, C., Goyat, S., Yaniv, M., and Thierry, F. (1997) Expression of the papillomavirus E2 protein in HeLa cells leads to apoptosis. EMBO J. 16, 504–514. 7. Sanchez-Perez, A. M., Soriano, S., Clarke A. R., and Gaston, K. (1997) Disruption of the human papillomavirus type 16 E2 gene protects cervical carcinoma cells from E2F-induced apoptosis. J. Gen. Virol. 78, 3009–3018. 8. Kanda, T., Sullivan, K. F., and Wahl, G. M. (1998) Histone-GFP fusion protein enables sensitive analysis of chromosome dynamics in living mammalian cells. Curr. Biol. 8, 377–385.

432

Kowalczyk et al.

9. Webster, K., Taylor, A., and Gaston, K. (2001) Oestrogen and progesterone increase the levels of apoptosis induced by the human papillomavirus type 16 E2 and E7 proteins. J. Gen. Virol. 82, 201–213. 10. Graessmann, M. and Graessmann, A. (1983) Microinjection of tissue culture cells. Methods Enzymol. 101, 482–492. 11. Stephens, D. J. and Allan, V. J. (2003) Light microscopy techniques for live cell imaging. Science 300, 82–86. 12. Nuñez, G., Benedict, M. A., Hu, Y., and Inohara, N. (1998) Caspases: the proteases of the apoptotic pathway. Oncogene 17, 3237–3245.

Codon Optimization of Papillomavirus Genes

433

31 Codon Optimization of Papillomavirus Genes Martin Müller Summary Early and late genes of human and animal papillomaviruses show a codon composition seemingly unfavorable for expression in mammalian cells. It remains unclear how the viruses manage to achieve high levels of late gene expression during the viral life cycle. One possible solution could be that the availability of certain t-RNAs changes with progressing stages of cellular differentiation. Previous studies have demonstrated that modification of codon usage of papillomavirus late (L1 and L2) and early genes (E7) can overcome poor expression of these proteins both in transient and in stable expression systems. This was shown not only for human but also for plant cells. Two strategies can be employed to alter codon usage: elimination of only those codons that are rarely used in a particular expression system, or exchange of all possible codons by the ones most frequently used. Currently, there are two protocols for codon modification—a template-less polymerase chain reaction (PCR)-based protocol, in which very long overlapping oligodeoxynucleotides are used in an overlap-extension reaction, or a ligase chain reaction, in which shorter oligodeoxynucleotides are fused together after an annealing procedure. Both methods are presented and discussed.

1. Introduction Earlier attempts to express various papillomavirus proteins in mammalian cells, but also in vivo, using transient and stable transfection methods, were met with only limited success. This became particularly obvious when studying the functions of the papillomavirus structural genes L1 and L2, for which, in contrast to the nonstructural papillomavirus genes, high protein yield had been expected during the viral life cycle when virions are assembled. However, it seems that the papillomaviruses are able to keep expression of the structural genes under very tight control by mechanisms involving not only transcription but also translational regulation. In order to produce and study the viral capsids, two different strategies have been followed to increase the amount of the L1 and L2 protein in mammalian cells. First, negative regulatory From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

433

434

Müller

elements present on the L1 and L2 mRNA have been identified and subsequently eliminated (1–7). These elements control posttranscriptional events such as mRNA stability and nuclear export of the late messages. However, although inactivation of these elements led to a somewhat increased accumulation of the L1 and L2 proteins, the overall protein yield had still not been satisfactory. The second strategy was the use of heterologous viral expression systems—in particular, the use of recombinant vaccinia viruses (VV) (see Chapter 33) and recombinant semliki forest viruses (SFV) (8,9). These expression systems allow high-yield production of L1 assembled into virus-like particles and the investigation of intracellular functions of the L1 and L2 gene. It is not fully understood why L1 and L2 can be expressed by VV or SFV. A possible explanation is that in both systems, the genes are transcribed in the cytoplasm, circumventing nuclear events such as regulators of nuclear export. The major limitation for the use of these expression systems is the cytopathic effect of the recombinant VV and SFV preventing expression over longer periods of time. In addition, there is indeed evidence that heterologous viral proteins interfere with the analysis of L1 and L2 functions. In the past, modification of the codon usage of several genes has been very successful to increase protein yield in a number of expression systems, including mammalian cells (10–13). Probably the most acknowledged example is the optimization of the green fluorescent protein (GFP) expression, without which the widespread use of GFP as a reporter gene in mammalian cells would not have been possible (12,13). What is codon optimization, or better, codon improvement (it is difficult to define an “optimal” codon composition)? Two strategies have been used: (1) replacement of extremely rare codons from the mRNA by codons frequently used in the respective expression host, or (2) replacement of all possible codons by codons most frequently used in the chosen cell system. Codons occurring at high frequencies can be identified for many different organisms by the use of various databases (14). Preferably, one would opt for codons frequently found in proteins expressed at high levels, but such information is as yet available for yeast cells only. The rationale in codon optimization is that low concentrations of certain t-RNAs are a limiting factor for protein translation (15,16). Eliminating only the rarest codons of a given gene can be achieved by standard mutagenesis techniques. For example, by this strategy about one-fifth of the codons of the GFP gene have been optimized for successful expression in mammalian cells. The more radical procedure of optimizing all possible codons (usually up to two-thirds of all codons) requires de novo synthesis of the complete coding sequence, but the results obtained by this method seem to be generally more promising. Both methods have been used to improve expression of early and late papillomavirus genes (17–22) (see Chapter 32). The results in particular have been very convincing


435

Fig. 1. Expression of codon-optimized L1 genes in human cells. Different mammalian expression constructs encoding codon-optimized human papillomavirus (HPV)16 L1 (L1h) or a C-terminally truncated L1h fused to unmodified HPV-16 E7 were introduced into the human cell line 911 by transient transfection (calcium-phosphate precipitation). Expression of the L1 protein was analyzed by Western blotting and indirect immunofluorescence using an L1-specific monoclonal antibody (Camvir-I). Interestingly, the expression of the L1/E7 fusion constructs is comparable to the L1h construct, although the E7 portion shows a codon usage which is not favorable for human cells. Note that the L1hE7 protein lacks a nuclear localization signal due to the truncation of L1. In two constructs, a SV40 NLS was inserted between L1h and E7.

in the case of the structural genes (see Fig. 1) but also of the early protein E7. The use of optimized genes allowed the development of nucleic acid-based vaccines, the production of infectious papillomavirus pseudovirions and the analysis of intracellular functions of the viral proteins (20,23). Despite the success of the newly generated papillomavirus genes, the mechanism(s) by which codon optimization leads to increased protein accumulation remains elusive. The radical optimization of the coding sequence not only alters the codon usage but very likely wipes out all or most of the regulatory elements present on the mRNA. Additionally, altering the mRNA sequence can have a strong impact on the mRNA stability (17). For some of the L1 genes, it was demonstrated, however, that codon optimization improved protein expression mainly by enhancing translational efficiency, as it would be expected (15,16,20). Nevertheless, some observations also clearly indicate that the results from optimization may be difficult to predict and that the reason for improved expression

436

Müller

might not be as simple to explain. For example, when testing L1 optimized for expression in human (L1h) or plant (L1p) cells with the original L1 gene (L1ori), we found that both modified genes show a strongly improved protein yield in mammalian cells compared to L1ori. However, surprisingly, only the L1h gene was expressed successfully in plant cells (17). In addition, L1ori and L1h can both be expressed in insect cells using recombinant baculoviruses, despite their contrasting codon composition. Nevertheless, it is also clear that in most instances expression of the optimized papillomavirus genes proved to be clearly superior compared to the unmodified genes, and many published and on-going studies have only been possible using expression-optimized genes. 2. Materials 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19.

PWO polymerase and buffer (Roche). Oligodesoxynucleotides. dNTPs. T4-DNA ligase and buffer. Restriction enzymes. Electrocompetent Escherichia coli DH5α and E. coli SURE bacteria. Agarose gel electrophoresis equipment. Themocycler. DNA-sequencing equipment. Pfu ligase and buffer. Oligo-desoxynucleotide (sets). T4 polynucleotide kinase and buffer (NEB). ATP. Cloning vector (such as pUC or pBluescript). Isopropyl-β-D-thiogalactopyranoside (IPTG), X-Gal. Expression vector. Transfection reagent, mammalian cells. Glass wool or QIAquick Gel extraction kit. Site-directed mutagenesis kit (e.g., Quickchange Stratagene).

3. Methods 3.1. Design of the Synthetic Gene

3.1.1. Codon Optimization It is recommended to replace as many codons in the gene as possible with codons most frequently used in the species for which expression is to be optimized (see Note 1). The most frequently used codons for humans are shown in Table 1. For other species, the reader should refer to the Kazusa codon-usage database (http://www.kazusa.or.jp/codon). Generation and editing of a codon-


437

Table 1 Codons Frequently Used in Human Genes Ala: GCC, GCU Arg: AGG, CGC, CGG Asn: AAC Asp: AAC Cys: UGC Gln: CAG Glu GAG Gly: GGC, GGG His: CAC Ile: AUC

Leu: CUG Lys: AAG Met: AUG Phe: UUC Pro: CCC Ser: AGC, UCC Thr: ACC Trp: UGG Tyr: UAC Val: GUG, GUC

The most frequently found codons of human genes are shown. For five of the amino acids, more than one codon can be used in codon optimization, because these codons are found at comparable frequencies.

optimized sequence can be achieved by the use of a word processor. The protein sequence (in the three-letter format) is converted to a DNA sequence by replacing the amino acids with their respective codon(s). Alternatively, Web tools are available that aid in the generation of synthetic genes (http:// genome.nci.nih.gov/publications/papilloma_ADAP.html). Because there are multiple possible codons for most of the amino acids (e.g., the three codons: AGG, CGC, CGG for arginine are found with similar frequencies in human genes), it is recommended to alternate codons whenever possible (see Note 1). This will result in a less monotonous sequence of the resulting gene and thereby avoid artifacts by false priming during the polymerase chain reaction (PCR) reactions.

3.1.2. Kozak Sequence In addition to the alteration of the codon sequences, it is recommended to insert a Kozak consensus sequence, (GCC)GCCA/GCCAUGG flanking the AUG initiation codon, as this has been shown to improve the efficiency of translation initiation (24).

3.1.3. Introduction of Unique Restriction Endonuclease Sites For cloning purposes, the synthetic genes are usually flanked by unique sites for restriction endonucleases. Current methods of gene synthesis are prone to high rates of error, resulting in point mutations and deletions. To aid removal

438

Müller

Fig. 2. Polymerase chain reaction (PCR)-based gene synthesis. Synthetic genes are designed with altered codon usage and contain additional unique recognition sites for restriction endonucleases (A, Re). Long oligodeoxynucleotides (ODN, 80–90mers) are synthesized, covering both sense and anti-sense strands (B). The ODNs on each strand are spaced by about 60 nt; ODNs of sense and anti-sense strand have an overlap of approx 20–22 nt. ODNs are annealed and the gaps are filled in a PCR. Using this, two or more fragments of the synthetic gene are produced (C). These fragments overlap each other and can be used as template in a combined primer-extension/PCR (overlap extension method; C–E).

of possible mutations, it is recommended to introduce additional unique recognition sites for restriction endonucleases every 200–300 basepairs (bp). This can be mostly achieved by altering single nucleotide positions, leaving the overall codon preference unaltered. It is further advisable to synthesize longer genes in fragments of 400–500 bp, which can be later assembled into a fulllength gene.

3.2. PCR-Based Method for Gene Synthesis 3.2.1. Oligodeoxynucleotides A set of overlapping oligodesoxynucleotides (ODN) spanning the complete synthetic gene needs to be synthesized (25). The ODNs are typically 80 (up to 100) nucleotides (nt) in length (see Note 2). The ODNs on each strand are spaced at approx 60 nt; ODNs of sense and antisense strands have an overlap of approx 20–22 nt, as shown in Fig. 2. Purity and quality of the ODNs are crucial for avoiding artifacts during the gene synthesis.


439

3.2.2. Gene Assembly by PCR 1. Four to eight ODNs can be used simultaneously in a single PCR as follows: 1 mM dNTP mix, 1X reaction buffer, 50 pmol of each ODN, and 2 U PWO polymerase, in a total volume of 50 µL. 2. The reaction is carried out in a thermocycler using the following settings: 96°C for 5 min, followed by 30 cycles of 1 min at 96°C, 1 min at 52°C, 1 min at 72°C, then a final step of 72°C for 10 min.

3.3. Alternative Method for Gene Synthesis: Ligase Chain Reaction-Based Method The PCR-based procedure for de novo gene synthesis involves the use of very long oligodeoxynucleotides in combination with one or more rounds of PCR amplification, eventually leading to the accumulation of point mutations and small deletions as a result of false priming events. Especially for humanized genes, which typically possess a very high GC content (up to 80%), this can be a considerable problem. The quality of the product very much depends on the integrity and quality of the long ODNs. The ligase chain reaction (LCR) based method allows gene synthesis with high accuracy, using much shorter ODNs that are easier to produce (26,27). In this strategy, both strands of the gene are completely synthesized in the form of ODNs (usually 40-mers). Again, ODNs derived from sense and antisense strands are staggered to allow complete annealing of the strands (see Fig. 3). Nicks between the annealed ODNs are sealed by the a thermostable DNA ligase, allowing the annealing process to be carried out at high temperatures, which minimizes the risk of artifacts due to secondary structures present in the ODNs.

3.3.1. Oligodeoxynucleotides A set of ODNs completely covering both sense and antisense strands is synthesized (see Note 2). ODNs from the sense strand overlap those of the antisense strand by 20 nt, as shown in Fig 3.

3.3.2. Phosphorylation of Oligodeoxynucleotides 1. To allow ligation of the annealed ODNs, all but the two 5' outermost ODNs of both strands need to be phosphorylated. For this, a mixture containing equal amounts of the ODNs is treated with ATP and T4 polynucleotide kinase in the following reaction: 1X polynucleotide kinase reaction buffer, 100 U T4 polynucleotide kinase, 2 µM each ODN, 1 mM ATP, in a total volume of 100 µL. 2. The reaction is carried out for 2 h at 37°C. 3. The kinase is then inactivated for 20 min at 65°C. 4. The phosphorylated ODNs are precipitated with ethanol and finally dissolved together with the two unphosphorylated ODNs in 50 µL H2O (4 µM each ODN).

440

Müller

Fig. 3. Long chain reaction (LCR)-based gene synthesis. In contrast with the polymerase chain reaction (PCR)-based method, shorter synthetic oligodesoxynucleotides (ODNs) (35–45 mers) are completely covering the sense and anti-sense strands (A). ODNs from the sense strand overlap with those of the anti-sense strand, allowing the annealing of the ODNs to the synthetic gene fragment. Nicks between the annealed ODNs are sealed by a heat-stable DNA ligase (B). Using this, two or more fragments of the synthetic gene are produced (C). These fragments overlap each other and can be used as template in a combined primer-extension/PCR (overlap extension method; C–E).

3.3.3. Ligase Chain Reaction 1. The phosphorylated ODNs are annealed and ligated in the following reaction: 2 µM of each of the phosphorylated and the two unphosphorylated ODNs), 1X Pfu DNA ligase reaction buffer, 10 U Pfu DNA ligase, in a total volume of 100 µL. 2. The LCR is carried out in a thermocycler using the following settings: 96°C for 5 min, followed by 15 cycles of 30 s at 96°C, 1.5 min at 55°C, 90 s at 70°C, and a final cycle of 30 s at 96°C, 2 min at 55°C, and 10 min at 70°C.


441

3.4. Isolation of Assembled DNA Fragments Assembled gene fragments (from PCR- and LCR-based gene synthesis) are analyzed by agarose gel electrophoresis. Fragments are isolated from the gels and purified by the “freeze-squeeze” method, i.e., by freezing the gel fragment and subsequently spinning the DNA-containing solution through a glass-wool cushion. Alternatively, commercially available purification kits (e.g., QIAquick, Qiagen) can be used to recover the fragments. If the fragments cover the complete gene, restriction digests with subsequent cloning will be carried out. If the fragments cover only parts of the synthetic gene, they will be used in a combined primer extension/PCR (overlap extension method, as shown in Figs. 2 and 3C–E).

3.5. Generation of Full-Length Genes From Fragments For longer synthetic genes, both the PCR- and the LCR-based gene assembly protocol will produce overlapping gene fragments (see Fig. 3D). These fragments can be fused by cycles of denaturation, priming, and elongation (overlap extension). The resulting full-length product can be amplified by PCR. All steps are carried out in a single reaction. 1. Set up the following reaction: 0.2 mM dNTPs, 1X reaction buffer, 10–50 ng of each overlapping DNA fragment, 25 pmol of each primer, 2 U of PWO polymerase, in a total of 50 µL. 2. Carry out the reaction in a thermocycler using the following settings: 96°C for 5 min, followed by 30 cycles of 1 min at 96°C, 1 min at 55°C, and 90 s at 72°C, then a final step of 72°C for 10 min.

3.6. Cloning into Vectors Isolated, full-length fragments are digested with the appropriate restriction enzymes and inserted into a suitable cloning vector, which is then transformed into a suitable E. coli strain (see Note 3). Multiple clones of each fragment are selected and analyzed by DNA sequencing. Mutations can be eliminated by exchange of correct subfragments between different clones or by site-directed mutagenesis (see Note 4). Fully assembled genes are transferred into a suitable expression vector, and expression is analyzed by Western blotting and/or indirect immunofluorescence (see Note 5). 4. Notes 1. Gene design. Despite advances in molecular biology techniques, gene synthesis remains a laborious and also costly undertaking. It is therefore crucial to invest sufficient efforts in the design of the synthetic gene. At this early step, additional features can be easily introduced into the product at no extra costs. Later modifi-

442

2.

3.

4.

5.

Müller cations might be difficult due to the nature of some synthetic genes. In which vectors will the gene be used? Should there be the possibility of protein tagging or fusion, or the expression of protein domains in addition to the full-length protein? Is there a need for the addition of sequences 5' or 3' to the coding sequence, such as ribosome binding sites in the case of genes to be expressed in prokaryotes? Highly repetitive sequences (especially long CG stretches) should be avoided. Scan the novel gene for the presence of putative mRNA degradation signals or splice sites. Quality of ODNs. The quality of the ODNs is crucial to the procedure, especially if long ODNs (>70 nts) are used for PCR-based gene synthesis. Order highest quality ODNs, purified by high-performance liquid chromatography (HPLC). Mutations accumulating in defined positions might indicate that one or more ODNs are of bad quality. Recombination and stability. Synthetic genes typically show a rather monotonous sequence, with a number of repeats due to the strongly reduced codon usage. For cloning and propagation, it is therefore advisable to use E. coli strains devoid of recombination, such as E. coli SURE. Elimination of mutations. Most of the assembled DNA fragments will contain one or more mutations. It is advisable to sequence multiple clones of each fragment. If appropriate restriction sites are available, exchange of intact parts between the different clones can speed up elimination of mutations. There are various methods to eliminate point mutations and small deletions, including commercially available kits (e.g., Quickchange, Stratagene). Problems cloning synthetic genes. In our experience, synthetic genes are sometimes difficult to handle in downstream cloning procedures. For unknown reasons, some of the genes are rather resistant for subcloning into expression vectors. Alternative vectors or alternative cloning strategies might solve these problems.

References 1. Carlsson, A. and Schwartz, S. (2000) Inhibitory activity of the human papillomavirus type 1 AU-rich element correlates inversely with the levels of the elav-like HuR protein in the cell cytoplasm. Arch. Virol. 145, 491–503. 2. Kennedy, I. M., Haddow, J. K., and Clements, J. B. (1990) Analysis of human papillomavirus type 16 late mRNA 3‚Ä processing signals in vitro and in vivo. J. Virol. 64, 1825–1829. 3. Kennedy, I. M., Haddow, J. K., and Clements, J. B. (1991) A negative regulatory element in the human papillomavirus type 16 genome acts at the level of late mRNA stability. J. Virol. 65, 2093–2097. 4. Rollman, E., Arnheim, L., Collier, B., et al. (2004) HPV-16 L1 genes with inactivated negative RNA elements induce potent immune responses. Virology 322, 182–189. 5. Schwartz, S. (2000) Regulation of human papillomavirus late gene expression. Ups. J. Med. Sci. 105, 171–192.


443

6. Sokolowski, M., Tan, W., Jellne, M., and Schwartz, S. (1998) mRNA instability elements in the human papillomavirus type 16 L2 coding region. J. Virol. 72, 1504–1515. 7. Sokolowski, M., Zhao, C., Tan, W., and Schwartz, S. (1997) AU-rich mRNA instability elements on human papillomavirus type 1 late mRNAs and c-fos mRNAs interact with the same cellular factors. Oncogene 15, 2303–2319. 8. Day, P. M., Roden, R. B., Lowy, D. R., and Schiller, J. T. (1998) The papillomavirus minor capsid protein, L2, induces localization of the major capsid protein, L1, and the viral transcription/replication protein, E2, to PML oncogenic domains. J. Virol. 72, 142–150. 9. Zhou, J., Stenzel, D. J., Sun, X. Y., and Frazer, I. H. (1993) Synthesis and assembly of infectious bovine papillomavirus particles in vitro. J. Gen. Virol. 74 (Pt 4), 763–768. 10. Andre, S., Seed, B., Eberle, J., Schraut, W., Bultmann, A., and Haas, J. (1998) Increased immune response elicited by DNA vaccination with a synthetic gp120 sequence with optimized codon usage. J. Virol. 72, 1497–1503. 11. Uchijima, M., Yoshida, A., Nagata, T., and Koide, Y. (1998) Optimization of codon usage of plasmid DNA vaccine is required for the effective MHC class Irestricted T cell responses against an intracellular bacterium. J. Immunol. 161, 5594–5599. 12. Yang, T. T., Cheng, L., and Kain, S. R. (1996) Optimized codon usage and chromophore mutations provide enhanced sensitivity with the green fluorescent protein. Nucleic Acids Res. 24, 4592–4593. 13. Zolotukhin, S., Potter, M., Hauswirth, W. W., Guy, J., and Muzyczka, N. (1996) A “humanized” green fluorescent protein cDNA adapted for high-level expression in mammalian cells. J. Virol. 70, 4646–4654. 14. Nakamura, Y., Gojobori, T., and Ikemura, T. (2000) Codon usage tabulated from international DNA sequence databases: status for the year 2000. Nucleic Acids Res. 28, 292. 15. Gu, W., Li, M., Zhao, W. M., et al. (2004) tRNASer(CGA) differentially regulates expression of wild-type and codon-modified papillomavirus L1 genes. Nucleic Acids Res. 32, 4448–4461. 16. Zhou, J., Liu, W. J., Peng, S. W., Sun, X. Y., and Frazer, I. (1999) Papillomavirus capsid protein expression level depends on the match between codon usage and tRNA availability. J. Virol. 73, 4972–4982. 17. Biemelt, S., Sonnewald, U., Galmbacher, P., Willmitzer, L., and Müller, M. (2003) Production of human papillomavirus type 16 virus-like particles in transgenic plants. J. Virol. 77, 9211–9220. 18. Buck, C. B., Pastrana, D. V., Lowy, D. R., and Schiller, J. T. (2004) Efficient intracellular assembly of papillomaviral vectors. J. Virol. 78, 751–757. 19. Cid-Arregui, A., Juarez, V., and zur Hausen, H. (2003) A synthetic E7 gene of human papillomavirus type 16 that yields enhanced expression of the protein in mammalian cells and is useful for DNA immunization studies. J. Virol. 77, 4928–4937.

444

Müller

20. Leder, C., Kleinschmidt, J. A., Wiethe, C., and Müller, M. (2001) Enhancement of capsid gene expression: preparing the human papillomavirus type 16 major structural gene L1 for DNA vaccination purposes. J. Virol. 75, 9201–9209. 21. Liu, W. J., Gao, F., Zhao, K. N., et al. (2002) Codon modified human papillomavirus type 16 E7 DNA vaccine enhances cytotoxic T-lymphocyte induction and anti-tumour activity. Virology 301, 43–52. 22. Mossadegh, N., Gissmann, L., Müller, M., Zentgraf, H., Alonso, A., and Tomakidi, P. (2004) Codon optimization of the human papillomavirus 11 (HPV 11) L1 gene leads to increased gene expression and formation of virus-like particles in mammalian epithelial cells. Virology 326, 57–66. 23. Görnemann, J., Hofmann, T. G., Will, H., and Müller, M. (2002) Interaction of human papillomavirus type 16 L2 with cellular proteins: identification of novel nuclear body-associated proteins. Virology 303, 69–78. 24. Kozak, M. (1987) At least six nucleotides preceding the AUG initiator codon enhance translation in mammalian cells. J. Mol. Biol. 196, 947–950. 25. Jayaraman, K., Fingar, S. A., Shah, J., and Fyles, J. (1991) Polymerase chain reaction-mediated gene synthesis: synthesis of a gene coding for isozyme c of horseradish peroxidase. Proc. Natl. Acad. Sci. USA 88, 4084–4088. 26. Au, L. C., Yang, F. Y., Yang, W. J., Lo, S. H., and Kao, C. F. (1998) Gene synthesis by a LCR-based approach: high-level production of leptin-L54 using synthetic gene in Escherichia coli. Biochem. Biophys. Res. Commun. 248, 200–203. 27. Khorana, H. G. (1979) Total synthesis of a gene. Science 203, 614–625.

Generation of HPV Pseudovirions Using Transfection

445

32 Generation of HPV Pseudovirions Using Transfection and Their Use in Neutralization Assays Christopher B. Buck, Diana V. Pastrana, Douglas R. Lowy, and John T. Schiller Summary It has recently become possible to generate high-titer papillomavirus-based gene-transfer vectors. The vectors, also known as papillomavirus pseudoviruses (PsV), have been useful for studying papillomavirus assembly, entry, and neutralization, and may have future utility as laboratory gene-transfer tools or vaccine vehicles. This chapter outlines a simple method for production of PsV and their use in a high-throughput papillomavirus neutralization assay. The production method is based on transfection of a 293 cell line, 293TT, engineered to express high levels of SV40 large T antigen. The cells are co-transfected with codon-modified papillomavirus capsid genes, L1 and L2, together with a pseudogenome plasmid containing the SV40 origin of replication. Pseudogenome encapsidation within L1/L2 capsids is largely sequence independent, and plasmids entirely lacking PV sequences can be packaged efficiently, provided they are less than 8 kilobases in size. Non-infectious virus-like particles (VLPs) can also be produced after transfection of 293TT cells with L1 alone. Efficient purification of the PsV or VLPs is achieved by Optiprep (iodixanol) density gradient ultracentrifugation. Using these methods, it is possible to produce highly purified PsV with yields of at least 109 transducing units from a single 75-cm2 flask of cells. PsV encapsidating a secreted alkaline phosphatase (SEAP) reporter plasmid were used to develop a high-throughput in vitro neutralization assay in a 96-well plate format. Infection of 293TT cells is monitored by SEAP activity in the culture supernatant, using a highly sensitive chemiluminescent reporter system. Antibody-mediated PsV neutralization is detected by a reduction in SEAP activity. The neutralization assay has similar analytic sensitivity to, and higher specificity than, a standard VLP-based enzyme-linked immunosorbent assay (ELISA).

1. Introduction Several methods for in vitro production of papillomavirus virions or pseudovirions (PsV) have been reported. They include production in keratinocyte raft culture (see Chapters 12 and 14), in cultured monolayers of From: Methods in Molecular Medicine, Vol. 119: Human Papillomaviruses: Methods and Protocols Edited by: C. Davy and J. Doorbar © Humana Press Inc., Totowa, NJ

445

446

Buck et al.

mammalian cells after infection with recombinant vaccinia or Semliki Forest virus vectors expressing L1 and L2 (see Chapter 33), or in the test tube after reassembly of capsomeres in the presence of plasmid DNA (1–4). However, none of these strategies efficiently produces high titers of PsV. Because hightiter PsV carrying an easily scored marker gene were unavailable, papillomavirus neutralization assays have been laborious, both in terms of production of the infectious capsids and in the conduct of the neutralization assays. In this chapter, we provide a simple and flexible procedure for generating papillomavirus pseudovirions with titers in excess of 109 transducing units per mL. We have used PsV produced by this method to develop a simple highthroughput assay for detecting papillomavirus-neutralizing antibodies. The PsV production strategy outlined under Subheadings 3.1. and 3.2. is based on maximizing the production of the two PV capsid proteins, L1 and L2, together with a target reporter plasmid (pseudogenome), in mammalian cells. Because expression of L1 and L2 is normally very low in cultured mammalian cells, L1 and L2 genes with extensive codon modification (see Chapter 31), must be used to overcome negative regulatory features of the wild-type open reading frames (ORFs) (reviewed in ref. 5). These codon changes do not change the primary amino acid sequence of the proteins, but do lead to a large increase in capsid protein production. To generate high-copy-number pseudogenomes for packaging, an SV40 origin of replication (ori) is inserted into the target plasmid, and the pseudovirus is produced in cells that express high levels of SV40 large T antigen (LT). 293T is an adenovirus-transformed human embryonic kidney cell line that was transfected with the SV40 genome. However, this line expresses very low levels of LT because of a splicing bias in favor of small t antigen (6). We generated a subclone, designated 293TT, expressing high levels of LT, by stable transfection of 293T cells with an expression plasmid encoding a cDNA for LT. This line supports high-level replication of plasmids containing the SV40 ori. Cotransfection of 293TT cells with plasmids containing strong eukaryotic promoters driving L1 and L2, together with the pseudogenome plasmid containing the ori and a marker gene, results in high-level expression of the three components and generation of hightiter PsV. Alternatively, non-infectious L1 virus-like particles (VLPs) can be produced after transfection of 293TT cells with a plasmid containing a codonmodified L1 gene and the SV40 ori. PsV stocks can consist of simple crude extracts of detergent-lysed producer cells. However, for many applications it is desirable to separate PsV capsids from cell components. A simple and efficient scheme for papillomavirus PsV and VLP purification is presented under Subheading 3.3. It is based on separation of the capsids from cell debris and detergent by high-salt extraction followed by ultracentrifugation in an Optiprep (iodixanol) step gradient. Under


447

the conditions employed, Optiprep produces a combined velocity sedimentation and buoyant density gradient. It produces excellent separation of PV capids from cell components and also achieves partial separation of PsV from VLPs. Unlike CsCl, which is often used for gradient purification of virus particles, Optiprep has relatively low osmotic content and is nontoxic to cells at concentrations of up to 30% w/v (7). Although Optiprep is considered a relatively gentle ultracentrifugation medium, the capsids of most papillomavirus types are too fragile to withstand purification immediately after release from producer cells. It was therefore necessary to devise a method to “mature” the capsids into a more stable conformation. A method for maturing PV capsids by simple overnight incubation of the crude cell lysates at 37°C is presented under Subheading 3.2.2. Employing the production, maturation, and purification strategies reported herein, it is possible to generate high-titer mature PsV stocks with particle-toinfectivity ratios of less than 10, using green fluorescent protein (GFP) as a marker of infection. PsVs encapsidating a secreted alkaline phosphatase (SEAP) reporter plasmid were used to develop the in vitro neutralization assay presented in Subheading 3.4. Transduction of 293TT cells is monitored by SEAP activity in the culture supernatant, using a highly sensitive chemiluminescent reporter system. Antibody-mediated PsV neutralization is detected by a reduction in SEAP activity. This is the first PV neutralization assay to be adapted to a highthroughput 96-well plate format. A single 75-cm2 flask can produce sufficient SEAP PsV for conducting thousands of assays. The neutralization assay appears to be as sensitive as, but more specific than, a standard VLP-based enzyme-linked immunosorbent assay (ELISA), and requires similar operator effort to that of an ELISA. The assay should have utility in both vaccine and sero-epidemiology studies. 2. Materials 1. 293TT cells (8). 2. DMEM-10: Dulbecco’s modified Eagle’s medium (DMEM), 10% 56°C inactivated fetal calf serum (FCS), 1% nonessential amino acids, 1% Glutamax-I (Invitrogen). 3. 50 mg/mL Hygromycin B stock (Roche). 4. 0.05% trypsin/ethyenediamine tetraacetic acid (EDTA) (Invitrogen). 5. OptiMEM-I (Invitrogen). 6. Lipofectamine 2000 (Invitrogen). 7. Papillomavirus L1 or L1/L2 expression plasmid (8–11). 8. “Pseudogenome” reporter plasmid (e.g., pYSEAP [Fig. 1], 10). 9. Siliconized pipet tips (VWR). 10. Siliconized 1.5-mL screw-cap tubes (Fisher, Cat. No. 05-541-63).

448

Buck et al.

Fig. 1. Maps of plasmids p16L1-GFP and pYSEAP.


449

11. Siliconized 1.5-mL flip-cap tubes (Fisher, Cat. No. 0554131). 12. Dulbecco’s phosphate-buffered saline (PBS) (Invitrogen, Cat. No. 14287-080). 13. 10% Brij58 (polyoxyethylene 20 cetyl ether) (Sigma) in Dulbecco’s PBS (stable for 2 mo at 4°C). 14. Lysis buffer: (Dulbecco’s PBS supplemented with 0.25% Brij58, 9.5 mM MgCl2, 0.1% Benzonase [Sigma], and 0.1% Plasmid Safe [Epicentre]). Prepare just prior to use. 15. 5 M NaCl. 16. 46% Optiprep: for 100 mL (77 mL 60% Optiprep (Sigma D1556), 10 mL 10X PBS, 12.5 mL 5 M NaCl, 45 µL 2 M CaCl2, 25 µL 2 M MgCl2, 210 µL 1 M KCl). Protect from light and store at room temperature for up to 2 mo. 17. Dulbecco’s PBS/0.8 M NaCl: for 100 mL (77 mL sterile water, 10 mL 10X PBS, 12.5 mL 5 M NaCl, 45 µL 2 M CaCl2, 25 µL 2M MgCl2, 210 µL 1 M KCl). Filtersterilize; stable for at least 2 mo at room temperature. 18. Polyallomer ultracentrifuge tubes (Beckman). 19. Swinging bucket ultracentrifuge rotor rated for >200,000g (e.g., SW55ti). 20. Sodium dodecyl sulfate (SDS)-polyacrylamide gel electrophoresis (PAGE) equipment. 21. Neutralization/growth media: DMEM without phenol red (see Note 1), 10% FCS (heat inactivated at 56°C), 1% MEM nonessential amino acids (Invitrogen, Cat. No. 11140-050), 1% HEPES (Invitrogen, Cat. No. 15630-106), 1% Glutamax-I (Invitrogen, Cat. No. 35050-061), 1% antibiotic-antimycotic (Invitrogen, Cat. No. 15240-062). 22. Dilution plates: untreated, sterile, U-bottom 96-well (Corning Costar, Cat. No. 3788). 23. 96-Well flat-bottom tissue-culture treated plates (Corning Costar, Cat. No. 3596). 24. Assay plates: Optiplate-96, white for luminescence, isotopic, and fluorescence (Perkin Elmer, Cat. No. 6005299). 25. Multichannel pipettor. 26. Sterile reservoir for use with multichannel pipettor. 27. Polystyrene 15- or 50-mL conical tubes. 28. Positive controls for neutralization assay: either known neutralizing antibodies or heparin 16,000 kD from porcine intestinal mucosa (Sigma H-4784) (12). 29. Chemiluminescent Great Escape SEAP detection kit (BD Bioscience/Clontech, Cat. No. 631701). 30. Microplate Luminometer (e.g., MLX luminometer, DYNEX).

3. Methods The methods described below outline (1) culture and transfection of 293TT cells, (2) harvest of PsV, (3) purification of PsV by ultracentrifugation through an Optiprep gradient, and (4) use of PsV in neutralization assays.

3.1. Culture and Transfection of 293TT Cells PsV are produced in 293TT cells by transient co-transfection of plasmids encoding the papillomavirus structural genes, L1 and L2, together with a

450

Buck et al.

reporter plasmid (pseudogenome). The protocol uses production of an HPV-16 PsV encoding SEAP reporter as an example. See Notes 2 and 3 for information about production of VLPs and other PsV types.

3.1.1. Thawing 293TT Cells 293TT are cultured in DMEM-10. To thaw 293TT cells, place the cells directly into a 150-cm2 flask with 25 mL of DMEM-10 supplemented with an additional 10% FCS. It is not necessary to spin the cells out of their freezing medium. Like other types of 293 cell lines, 293TT do not adhere tightly. It may take as many as 3 d for the cells to attach after thawing.

3.1.2. Passaging 293TT Cells Split 293TT cells 1:5 to 1:20 when they reach 80–90% confluence. Detach cells by gently rinsing the flask once with several mL of trypsin, followed by a 5–10 min incubation in 2 mL of fresh trypsin in a humidified 37°C incubator. It is important to trypsinize the cells thoroughly, since insufficient trypsinization can lead to shredding of cell clumps. Inactivate trypsin by resuspending the cells in DMEM-10 and split directly into a fresh flask. After recovery from the thaw, DMEM-10 can be supplemented with 400 µg/mL hygromycin B to promote maintenance of T antigen expression. Although 293TT cells can typically be passaged for several months without alteration of PsV production or titration characteristics, an early passage should be frozen in aliquots for longterm storage.

3.1.3. Transfection Use Invitrogen’s Lipofectamine2000 essentially according to the package insert. 1. Preplate 10 million 293TT cells in 20 mL of DMEM-10 (no hygromycin or antibiotics) in a 75-cm2 flask 16 to 24 h prior to transfection. 2. Mix 13 µg each of p16L1-GFP, p16-L2h, and pYSEAP (see Notes 2 and 3) with 2 mL of OptiMEM-I. 3. In a separate tube mix 85 µL of Lipofectamine 2000 with 2 mL of OptiMEM-I. 4. Incubate the two mixtures separately at room temperature for 10 to 30 min, then combine and incubate for at least an additional 20 min. 5. Add the resulting lipid/DNA complexes directly to the preplated cells. It is not necessary to change medium prior to transfection. 6. Incubate the cells with the lipid/DNA complexes for 4 to 6 h, then remove the complexes and add fresh DMEM-10 prewarmed to 37°C. Add the fresh DMEM-10 to the top of the flask to avoid dislodging cells. Incubating cells overnight in lipid/DNA mix results in significant cytotoxicity without much improvement in transfection efficiency.


451

From this point forward, treat cells and all their products like an infectious agent (see Note 4). Incubate the transfected cells overnight, then split 1:2 or 1:3, depending on cell density. Use bleach or 70% ethanol to disinfect all plasticware before it exits the hood.

3.2. Harvest and Maturation PsV are released from 293TT cells by detergent lysis. Although the PsV are infectious immediately after release, pseudovirions must be given time to mature prior to purification. Maturation is accomplished by simple overnight incubation of the cell lysate at 37°C. The matured PsV is solubilized by addition of sodium chloride to the lysate, which allows clarification of the lysate by low-speed centrifugation.

3.2.1. Collect Cells Collect producer cells by trypsinization about 44 h after transfection (see Note 5). If there are many floating cells, collect them by centrifugation and combine with trypsinized cells. Resuspend the cells in 10 mL of DMEM-10 and count viable cells by trypan blue exclusion. An initial transfection of a T-75 flask should yield about 50 million cells. Spin down the cells and discard the supernatant. Transfer the cells into a siliconized (see Note 6) 1.5- or 2.0-mL screw-cap tube using 2 × 0.5 mL of DPBS. Spin down the cells and discard supernatant.

3.2.2. Cell Lysis and Capsid Maturation Suspend cells at approx 100 million cells per mL in lysis buffer. Note that the cell pellet occupies nearly a third of the final volume. For example, add about 400 µL of lysis buffer to a cell pellet of 60 million cells. Incubate cell lysate at 37°C for at least 16 h (see Note 7). Mix the tube by inversion occasionally, particularly during the first 2 h of the incubation.

3.2.3. Salt Extraction Note that Optiprep gradients (see Subheading 3.3.) must be allowed to diffuse at least an hour prior to performing salt extraction. 1. Chill the matured lysate on ice for 5 min. 2. Bring salt concentration up to 850 mM by adding 0.17 volume of 5 M NaCl. Incubate on ice for 10 to 20 min. If desired, a sample of salt-treated lysate can be withdrawn for titering (see Note 10). Crude stock must be diluted at least 1:500 to avoid detergent toxicity to the target cells. 3. Clarify the salt lysate by spinning for 15 min at 2000g in a refrigerated microcentrifuge.

452

Buck et al.

4. Transfer the supernatant onto an Optiprep gradient (see Subheading 3.3.) or see Note 8 for alternative purification methods. 5. Additional PsV can be recovered by resuspending the pellet with a few hundred µL of cold Dulbecco’s PBS/0.8 M NaCl, spinning again for 15 min, then adding the wash to the top of the Optiprep gradient.

3.3. Optiprep Purification PsV is purified by ultracentrifugation through an Optiprep step gradient. This section also outlines methods for biochemical analysis of the purified PsV.

3.3.1. Preparation of Optiprep Gradients Use DPBS/0.8 M NaCl to dilute 46% Optiprep stock (see Subheading 2.) to 27%, 33%, and 39%. Use 50-mL conical centrifuge tubes to allow easier syringe draws (discussed later). Pour Optiprep gradients in thin-wall polyallomer 5-mL tubes (e.g., Beckman 326819) by underlaying (27%, then 33%, then 39%) 1.4-mL steps using a 3-mL syringe fitted with a long needle. If necessary, pour a balance gradient. Allow the gradients to diffuse at room temperature for 1 to 4 h.

3.3.2. Ultracentrifugation Gently layer clarified cell lysate and wash (see Subheading 3.2.3.) onto the linearized gradient. The tubes should be full and the tubes/buckets should be balanced to within ±5 mg. Spin for 3.5 h, 16°C, at 234,000g in an SW55ti rotor. Set the acceleration and deceleration to “slow.” Too rapid an acceleration/deceleration may stir the gradients. Other types of rotors can be used successfully—for example, SW40.1Ti at 200,000g for 4.75 h or SW32 at 125,000g for 5.75 h.

3.3.2. Fraction Collection The L1 band may be faintly visible as a uniform light gray layer a little over a third of the way up the gradient. Collect gradient fractions by puncturing the bottom of the tube slightly off center with a 26-gage syringe needle. Collect fractions in siliconized microcentrifuge tubes. Collect the first approx 750 µL as one fraction, then collect 6- to 8-drop (approx 250 µL) fractions up to fraction 10. Discard the top approx 2 mL of the gradient.

3.3.3. Screening Fractions Because L1 is the major protein present in core fractions of the gradient (Fig. 2), fractions containing PsV can be identified rapidly by SDS-PAGE minigel analysis of 5 µL of each fraction. The colloidal coomassie reagent Microwave Blue (Protiga) offers a rapid and sensitive method for staining SDS-


453

Fig. 2. Sodium dodecyl sulfate-polyacrylamide gel electrophoresis analysis of Optiprep fractions.

PAGE gels. Fractions containing significant amounts of L1 (55 kD) should be pooled, aliquoted, and frozen at –80°C. Peak L1 content is generally found between fractions 4 and 8. SDS-PAGE analysis of bovine serum albumin standards ranging from 2 µg to 50 ng, or BCA protein assay (Pierce), can be used to quantitate L1 yield. Overall L1 yield should be about 200 µg of L1 from a 75-cm2 flask transfected with p16L1 GFP. Notes 9 and 10 discuss alternative methods for screening fractions.

3.4. Neutralization Assay The methods described below outline (1) titration of the SEAP-PsV stock, (2) luminometry to detect SEAP production, and (3) determination of the neutralization titer of test sera. The PsV used for this assay encapsidates a reporter plasmid, pYSEAP, encoding SEAP. When PsV infect 293TT cells, the pYSEAP reporter plasmid, which carries an SV40 ori, is replicated to high copy number. This leads to high-level production of alkaline phosphatase that is secreted into the culture medium, and so can be easily assayed. Antibodymediated neutralization of the PsV results in a corresponding reduction in SEAP expression.

3.4.1. Titration of SEAP-PsV Stocks Before assaying for neutralizing activity of test sera, it is important to titrate the PsV stock to determine the inoculum that will be used in each assay. The goal of the titration is to determine the minimum amount of PsV required to

454

Buck et al.

give a robust signal in the SEAP assay (see Subheading 3.4.2.) that is well above background but within the linear range of the assay. Typically, this falls in a range between 30 and 100 relative light units (RLUs) in the absence of neutralizing antibodies, with a background of no more than 1 RLU when the PsV is maximally neutralized with the positive control antibody or heparin. The method to titrate the stock is as follows: 1. 2. 3. 4. 5.

6. 7.

8. 9. 10.

11. 12. 13. 14.

Calculate the number of plates needed to titer the PsV. Trypsinize 293TT cells and suspend in neutralization/growth medium. Count the cells and preplate 2–5 h before the PsV are added (see Note 11). Dilute the cells to 300,000/mL in the neutralization/growth medium and place in a sterile reservoir. Using a multichannel pipettor, deliver 100 µL of cell suspension to each of the internal wells of a 96-well tissue-culture treated plate. To avoid evaporation, do not use external wells and instead fill surrounding wells with 120–150 µL of medium with phenol red (Table 1). Replace cover and place cells at 37°C in an incubator until ready to add the PsV. Using siliconized tubes and tips, make serial dilutions of the PsV. Each dilution is tested in triplicate, and enough should be diluted for at least six wells—three with and three without positive neutralization control. Depending on the papillomavirus type, appropriate dilution ranges between 1:300 and 10,000. Place 80 µL of the diluted PsV into the wells of the dilution plates. To the wells that will have untreated PsV, add 20 µL of neutralization/growth medium. Dilute the positive neutralization antibody (or heparin) such that it is fivefold more concentrated than its known 95% neutralizing dilution. For example: V5 (anti-HPV16 monoclonal) at 1:250,000 5B6 (anti-bovine papillomavirus [BPV] monoclonal) at 1:25,000 Rabbit anti-VLP polyclonal sera at 1:10,000 to 1:1,000,000 Heparin H-4784 at 1 mg/mL Dilute the antibody another fivefold by adding 20 µL of diluted antibody (above) to triplicate wells containing 80 µL of diluted PsV. Once the PsV and positive neutralization control(s) are combined, place on ice for 1 h. Add the whole mixture to the preplated cells. Return to the incubator for 72 h. The medium should not be replaced during these 72 h (see Note 12).

3.4.2. Chemiluminescent Detection of Secreted Alkaline Phosphatase For this section of the protocol use a multichannel pipettor when transferring liquids from one plate to the other. Make up kit reagents as indicated below and transfer to a reservoir so you can also use a multi-channel pipettor for those steps. Although SEAP activity can be detected colorimetrically, chemi-


455

Table 1 Schematic for a 96-Well Plate for Titering PsV*

*Striped cells should be filled with 120 µL of medium with phenol red to avoid evaporation from inner wells. HPV, human papillomavirus.

luminescent methods, such as the one described here, are generally preferable since they offer a much higher signal-to-noise ratio (see Note 13). 1. After the 72 h incubation, lightly shake plates to obtain a homogeneous sample of the supernatant. 2. Transfer 50 µL of supernatant to the corresponding well of a sterile 96-well polystyrene plate 3. Spin the plate at 1000g for 5 min. 4. Use Great Escape reagents according to manufacturer’s instructions. Briefly: 5. Add 45 µL of 1X dilution buffer directly into wells of a white optiplate-96 assay plate.

456

Buck et al.

6. Transfer 15 µL of clarified supernatant to the plate, cover with plastic coverfoil, and incubate 30 min at 65°C. 7. Incubate plates on ice 2–5 min. 8. Add 60 µL of room temperature 1X assay buffer and incubate at room temperature for 5 min. 9. Dilute chemiluminescence substrate and add 60 µL to each well. 10. Incubate at room temperature for 20 min. 11. Read on MLX Microplate Luminometer (Dynex Technologies) set at Glow-Endpoint 0.20 s/well RAW Data Handling Average readings at 20 min after adding substrate.

The relative light units (RLUs) obtained from triplicate samples should not vary by more than 10 or 15%. If they vary more than that, check the notes section to try to troubleshoot the problem.

3.4.3. Neutralization Assay Once the PsV has been titrated, test sera (see Notes 14 and 15) can be assayed to determine endpoint neutralization titers. To monitor inter-assay variability, the following controls should be included for each plate: (1) at least two wells of cells in neutralization/growth media without PsV or serum, (2) at least four wells of PsV-infected cells to which no antibody was added, (3) cells treated with PsV pre-incubated with a known serum, with at least 4 dilutions that span the 50% neutralizing titer that has been recorded in other experiments, and (4) cells treated with PsV pre-incubated with at least one dilution of a known non-neutralizing serum. See Table 2 for a typical arrangement of samples. 1. Preplate cells as described under Subheading 3.4.1. 2. Perform serial dilutions of the unknown sera (see Note 14) in sterile polystyrene plates. Be sure to take into account the 1:5 dilution of the serum once 20 µL of diluted serum are added to 80 µL of the PsV. 3. Prewet a sterile multichannel pipet reservoir with neutralization/growth medium to prevent sticking of PsV to the reservoir. 4. Dilute PsV in neutralization/growth medium to the concentration determined under Subheading 3.4.1. in a 15- or 50-mL polystyrene centrifuge tube. 5. Vortex briefly and decant into the prewetted reservoir. 6. Using siliconized tips and a multichannel pipettor, transfer 80 µL of the diluted PsV to each well of a 96-well dilution plate. If siliconized tips are unavailable, pipet diluted PsV up and down five times before delivering to the first row of the plates, then use the same tips for all other rows. 7. Add 20 µL of the test or control sera to the PsV and incubate on ice for 1 h. 8. Add the entire volume of the virus-antibody mixture to the corresponding wells of the plate using a multichannel pipettor. 9. Return the plate to the incubator for 72 h.


457

Table 2 Schematic Drawing for a Typical Setup for Determining Neutralizing Titer of Unknown Sera With Human Papillomavirus (HPV)-16 PsV*

*Striped wells should be filled with 120 µL of medium with phenol red to avoid evaporation from inner wells.

10. The medium should not be changed. 11. The supernatant is then assayed for presence of SEAP (see Subheading 3.4.2.). The titer is defined as the reciprocal of the highest dilution of serum that reduces the SEAP activity by at least 50% in comparison to the reactivity in the wells that received PsV but no antibody.

458

Buck et al.

If the same PsV stock is used to repeat the neutralization, then the 50% neutralizing titer should be the same, or vary by three- or fourfold. If the results vary by more than fourfold, the assay should be repeated a third time. Report the geometric mean titer of all assays performed. 4. Notes 1. The presence of phenol red tends to give a higher background in the chemiluminescent (and colorimetric—see Note 10) detection assay; therefore, medium without phenol red is preferred for the assays. 2. Maps of plasmids useful for PsV production are available at the website http:// ccr.cancer.gov/Staff/links.asp?profileid=5637. Care should be taken when re-transforming the plasmids, because the collection uses a wide variety of different drug-resistance markers. It is currently possible to generate BPV1, HPV-16, and HPV-18 PsV. Additional types are in development in various labs. For BPV1, it was possible to create a single large plasmid that drives expression of both L1 and L2 (pSheLL). The large size of the plasmid precludes self-packaging. For other papillomavirus types, L1 and L2 must be co-transfected on separate plasmids. In some instances, L1 and PsV yield can be increased substantially if plasmids expressing L1 under the control of the human elongation factor (EF)1α promoter also carry an SV40 origin of replication (e.g., p16L1-GFP, [11] and peL1fB, [10]). The protocol can be adapted to production of non-infectious L1-only VLPs simply by omitting the L2 and reporter plasmid components from the transfection step. 3. Although essentially any plasmid under 8 kb in size can be packaged by L1 and L2, PsV production efficiency varies with different reporter plasmids for reasons that are not fully understood. In general, smaller (approx 6 kb) plasmids expressing a reporter gene under control of EF1α promoter give better titer yield than plasmids expressing genes under control of CMV promoter. The presence of an SV40 ori on the target plasmid is not strictly required, but does typically augment titer yield by 5- to 10-fold. PsV carrying GFP reporter plasmids (e.g., pfwB, [11]) are convenient, since they can be easily titered by fluorescence-activated cell sorting (FACS). 4. Because PsV are capable of transferring foreign DNA, they should be handled using full biosafety level 2 precautions. It is important to note that the promiscuity of packaging by L1 and L2 can lead to generation of PsV with encapsidated fragments of cellular DNA, possibly including SV40 large tumor antigen, adenovirus oncogenes, or unknown oncogenes present in 293TT cells. Mature PsV can be inactivated by 70% ethanol, but, like many other types of non-enveloped virus particles, they are resistant to a wide spectrum of physical insults, including various detergents and proteases, prolonged heating at 56°C, 50 mM EDTA, and sodium chloride up to at least 1.5 M (unpublished results). PsV are probably also resistant to dessication (13). High-titer PsV should never be harvested using tip sonication, which can create potentially dangerous aerosols.


459

5. In most cases, substantial amounts of PsV are generated within 24 h posttransfection. However, maximum titer yield is typically approx 44 h posttransfection. Although the SV40 ori+ plasmid DNA content increases approx fivefold between 44 and 52 h, very little additional PsV titer appears during this period. Titer yield at 72 h is generally poor, presumably due to cell death triggered by over-replication of SV40 ori+ plasmid DNA. 6. It is important to use siliconized tubes and pipet tips, because PsV adsorb nonspecifically to polypropylene (14). Long-term storage of purified PsV or VLPs at 4°C can result in loss due to nonspecific sticking, even in siliconized and polystyrene tubes (see also Note 9). 7. During the overnight incubation of the cell lysate, pseudovirions “mature” into a more stable configuration. During maturation, disulfide bonds gradually form between L1 molecules, and the capsid becomes condensed and shows improved regularity in electron micrographs (11). Although the immature PsV are infectious immediately after cell lysis, they are too fragile to withstand purification. It is sometimes practical to use crude cell lysates as PsV stock. If purification is not necessary, the maturation step can be omitted and the stock can be frozen in aliquots after addition of salt (see Subheading 3.2.3.). It is important to perform the maturation step in high-density cell lysates. Cell lysates at densities of 20 million cells per mL or less suffer dramatic nonspecific protein aggregation. The nonspecific aggregation can lead to poor separation of capsids from cell debris in Optiprep gradients. Although it is possible to accelerate the maturation process by addition of oxidants (e.g., 5 mM oxidized glutathione), the accelerated maturation can lead to formation of irregular capsids (unpublished observation). 8. If ultracentrifugation equipment is unavailable, VLPs or PsV can be partially purified over disposible HiTrap Heparin HP affinity columns (Pharmacia Catalog No. 17 0406-01) (15). Papillomavirus capsids bind the columns in buffers with 0.3 M NaCl and elute at 0.8 M NaCl. Phosphate or MOPS can be used for buffering. Because papillomavirus virions are too large to enter pores in the crosslinked agarose matrix of the columns, the overall binding capacity of the columns can be as low as 100 µg of L1 per mL of column volume. Because many cellular proteins bind the columns, the degree of purification is quite poor compared to Optiprep gradient ultracentrifugation. It is possible to exchange PsV out of salt, detergent, or Optiprep-containing solutions using Amicon centrifugal filter devices (e.g., Millipore UFC81008). Finally, PsV stocks can be purified (or re-purified) by buoyant density centrifugation in a self-forming Optiprep gradient. PsV have an apparent density of about 1.20 g/mL in Optiprep, whereas empty papillomavirus capsids are closer to 1.25 g/mL. Separation of the two species can be achieved by mixing the PsV stock into 30% Optiprep/0.8 M NaCl followed by equilibrium centrifugation in an NVT65 rotor at 65,000 rpm for 4 h. 9. For some papillomavirus types (particularly BPV1), a significant percentage of L1 particles may lack encapsidated DNA. Empty capsids migrate to the higherdensity fractions (i.e., toward the bottom of the tube) relative to DNA-filled infectious PsV. For some applications, it may be desirable to achieve a lower

460

10.

11. 12. 13.

Buck et al. particle-to-infectivity ratio by discarding fractions containing empty capsids. If gradient SDS-PAGE gels are available (e.g., NuPAGE precast 4–12% Bis-Tris/ MOPS gels [Invitrogen]), it is often possible to discriminate between the two types of particles by visualization of histone-sized proteins (approx 15 kD) in fractions with significant PsV titer (Fig. 2). Another rapid method for screening fractions is to extract encapsidated DNA from 20 µL of each fraction using microcentrifuge silica columns (e.g., PCR Purification Kit [Qiagen]). In most instances, the extracted DNA can be easily visualized by agarose gel electrophoresis followed by staining with ethidium bromide or SYBR Green I (Sigma). Finally, the fractions can be titered for SEAP- or GFP-transducing activity individually (see Subheading 3.4.1. and Note 10). A drawback to titering the fractions is that they must be stored during the 2–3 d it takes to perform the titration. In order to avoid loss by nonspecific adsorption to the tube walls (see Note 6), the fractions (or aliquots of pooled fractions) should be stored at –80°C. Alternatively, it is possible to saturate nonspecific binding by adding 10% FCS to PsV stocks, allowing storage for up to a month at 4°C. Although mature PsV stocks typically suffer less than a 25% loss of titer during freezing, repeated freeze-thaw cycles should be avoided. PsV stocks are stable at –80°C for at least a year. If a GFP-expressing plasmid (e.g., p16L1-GFP [Fig. 1] or pfwB) was included in the transfection, it is possible to titer the stock using a FACS machine. For pfwB, titer yield should be at least 109 GFP-transducing units per 75-cm2 flask. GFPbased titration should be performed as follows: • Preplate 293TT cells in DMEM-10 in a 24-well plate at 1 × 105 cells in 0.5 mL per well. Incubate overnight. Alternatively, preplate 2 × 105 cells in 0.5 mL of DMEM-10 an hour or two in advance. Cells should be

Human Papillomaviruses: Methods and Protocols (Methods in Molecular Biology)

Human Papillomaviruses: Methods and Protocols

Human Papillomaviruses: Methods and Protocols

Human Embryogenesis: Methods and Protocols (Methods in Molecular Biology)

Yeast Protocols (Methods in Molecular Biology) (Methods in Molecular Biology)

Molecular Imaging: Methods and Protocols (Methods in Molecular Biology)

Molecular Chaperones: Methods and Protocols (Methods in Molecular Biology, v787)

Molecular Chaperones: Methods and Protocols (Methods in Molecular Biology 787)

Molecular Profiling: Methods and Protocols (Methods in Molecular Biology, 823)

Molecular Toxicology Protocols (Methods in Molecular Biology)

Bioinformatics. Methods and Protocols in Molecular Biology

Microchip Capillary Electrophoresis: Methods And Protocols (Methods in Molecular Biology) (Methods in Molecular Biology)

Epidermal Cells: Methods and Protocols (Methods in Molecular Biology) (Methods in Molecular Biology Series)

Cytoskeleton Methods and Protocols (Methods in Molecular Biology Vol 161)

Immunological Tolerance: Methods and Protocols (Methods in Molecular Biology)

Clostridium difficile: Methods and Protocols (Methods in Molecular Biology, 646)

Hepatocytes: Methods and Protocols (Methods in Molecular Biology 640)

Biofuels: Methods and Protocols (Methods in Molecular Biology 581)

Bioconjugation Protocols: Strategies and Methods (Methods in Molecular Biology)

Riboswitches: Methods and Protocols (Methods in Molecular Biology)

Bioinformatics Methods and Protocols (Methods in Molecular Biology)

Argonaute Proteins: Methods and Protocols (Methods in Molecular Biology 725)

Surface Plasmon Resonance: Methods and Protocols (Methods in Molecular Biology)

Neuropeptides: Methods and Protocols (Methods in Molecular Biology 789)

Microbial Toxins: Methods and Protocols (Methods in Molecular Biology 739)

Retinoids: Methods and Protocols (Methods in Molecular Biology)

Tumor Biomarker Discovery: Methods and Protocols (Methods in Molecular Biology)

Potassium Channels: Methods and Protocols (Methods in Molecular Biology)

Metabolomics: Methods and Protocols (Methods in Molecular Biology Vol 358)

Mast Cells: Methods and Protocols (Methods in Molecular Biology)

Tissue Microarrays: Methods and Protocols (Methods in Molecular Biology)

Human Papillomaviruses: Methods and Protocols (Methods in Molecular Biology)